Investigation of Neuronal Membrane Fusion Using

Transcrição

Investigation of Neuronal Membrane Fusion
Using Fluorescence Correlation Spectroscopy
Dissertation
for the award of the degree
“Doctor rerum naturalium”
of the Georg August University Göttingen,
submitted by
Wensi Vennekate
From Tongliaoshi, China
Göttingen 2012
ii
Members of the Thesis Committee:
Prof. Dr. Peter Jomo Walla (Referee)
Workgroup Biomolecular Spectroscopy and Single Molecule Detection
Max Planck Institute for biophysical Chemistry, Göttingen
and
Department Biophysical Chemistry
Institute for physical and theoretical Chemistry
Technische Universität Carolo-Wilhelmina zu Braunschweig
Prof. Dr. Reinhard Jahn (Referee)
Department Neurobiology
Max Planck Institute for biophysical Chemistry, Göttingen
Prof. Dr. Claudia Steinem
Institute for Organic and Biomolecular Chemistry
Georg August University, Göttingen
Date of oral examination: November 8th 2012.
I hereby declare that the Ph.D. thesis entitled “Investigation of Neuronal Membrane Fusion using Fluorescence Correlation Spectroscopy” has been written independently and with no other sources and aids than quoted.
Göttingen, September 30th, 2012.
Contents
1
Introduction
1
1.1
Neuronal exocytosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
1.2
The role of SNARE proteins in exocytosis . . . . . . . . . . . . . . . . . .
2
1.3
Calcium sensor protein: synaptotagmin-1 . . . . . . . . . . . . . . . . . .
6
1.4
NSF and α-SNAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9
1.5
Fluorescence correlation spectroscopy . . . . . . . . . . . . . . . . . . . .
11
1.5.1
Fluorescence correlation spectroscopy (FCS) . . . . . . . . . . . .
11
1.5.2
Fluorescence cross-correlation spectroscopy (FCCS) . . . . . . . .
14
1.5.3
Two-photon excitation (TPE) . . . . . . . . . . . . . . . . . . . . .
15
1.5.4
Förster resonance energy transfer (FRET) . . . . . . . . . . . . . .
17
Outline of this Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
19
1.6
2
Materials and Methods
21
2.1
Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
21
2.1.1
Chemicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
21
2.1.2
Phospholipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
21
2.1.3
Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
22
2.1.4
Buffers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
25
i
CONTENTS
2.2
3
2.1.5
Instruments, filters, columns and others . . . . . . . . . . . . . . .
29
2.1.6
Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
30
Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
30
2.2.1
Expression and purification of synaptotagmin-1 . . . . . . . . . .
30
2.2.2
Determination of the protein concentration . . . . . . . . . . . . .
34
2.2.3
Determination of Rhodamine Green concentration with UV absorption spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . .
35
2.2.4
SDS-PAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
36
2.2.5
Reconstitution of liposomes . . . . . . . . . . . . . . . . . . . . . .
37
2.2.6
Purification of mouse and rat SVs . . . . . . . . . . . . . . . . . .
38
2.2.7
Purification of chromaffin granules (CGs) . . . . . . . . . . . . . .
39
2.2.8
Two-photon confocal fluorescence microscopy setup . . . . . . .
39
Results
3.1
Cis- and trans-membrane interaction of synaptotagmin-1 . . . . . . . . .
43
3.1.1
Tethering assay based on FCS and FCCS . . . . . . . . . . . . . .
44
3.1.2
Cis- and trans-interaction of Syt1 to the acidic lipid membrane . .
46
3.1.3
The soluble C2AB domain of Syt1 is able to cluster liposomes at
a saturating concentration . . . . . . . . . . . . . . . . . . . . . . .
50
Syt1-SNARE interaction mediated membrane tethering . . . . . .
54
Characterization of mouse synaptic vesicles by average protein mass . .
57
3.2.1
Synaptic vesicle labeling with FM 1-43 dye . . . . . . . . . . . . .
59
3.2.2
Concentration determination of SVs using TP-FCS . . . . . . . . .
59
3.2.3
Background correction due to the dark FM fluorescence . . . . .
61
3.1.4
3.2
43
ii
CONTENTS
3.2.4
3.3
Comparing the mass of proteins per vesicle of mouse and rat SVs
α-SNAP inhibits SNARE-mediated membrane fusion but not partial
SNARE-zippering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3.1
3.3.2
4
5
64
Determination of CG docking by observing variations in diffusion time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
67
α-SNAP does not fully block CG docking . . . . . . . . . . . . . .
70
Discussion
73
4.1
The conflict of cis- and trans-membrane interaction of synaptotagmin-1 .
4.2
α-SNAP inhibits SNARE-mediated CG fusion, but does not fully block
4.3
62
73
CG docking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
78
Quantitative characterization of synaptic vesicles . . . . . . . . . . . . . .
80
Conclusions
81
Bibliography
83
List of Figures
101
List of Tables
105
Abbreviations and Symbols
107
Publication
111
iii
CONTENTS
iv
Chapter 1
Introduction
1.1
Neuronal exocytosis
Neuronal communication requires fast exocytosis of synaptic vesicles. Synaptic
vesicles (SVs) possess a large variety of trafficking proteins in different copy numbers and are filled with neurotransmitters in the lumen [119]. SVs function as transporters of neurotransmitters and release their neurotransmitters by rapid fusion with
the presynaptic plasma membrane triggered by calcium influx upon arrival of an action potential [62, 63, 139].
Filled SVs translocate to the active zone, in which SVs target the plasma membrane (docking) and prepare themselves for fusion (priming). The priming state involves protein-protein interaction as well as protein-lipid interaction (priming). Subsequently, they fuse with the plasma membrane initiated by the local increase of the
intracellular calcium concentration (fusion, see Figure 1.1) [62, 113, 117].
After fusion, new empty SVs are reformed from the plasma membrane by clathrin
coated vesicle fission, a process termed endocytosis [45]. The new SVs are recycled
by fusion with the early endosome and regenerated by budding from the early endosome [113]. The regenerated SVs are refilled with neurotransmitters triggered by
an electrochemical gradient (NT uptake) and proceed to a new round of the exo- and
1
CHAPTER 1. INTRODUCTION
endocytosis (Figure 1.1) [113].
Figure 1.1: Synaptic vesicle cycle [113]. Neurotransmission is mediated by fusion of synaptic
vesicles with the plasma membrane. During this process, neurotransmitters are
released from the vesicle lumen into the synaptic cleft. After fusion, new SVs are
reformed from the plasma membrane by endocytosis. Empty vesicles are recycled
from the early endosome by endosome fusion and regenerated by budding from the
early endosome. The vesicles are then filled with neurotransmitters (NT uptake)
and are available for neurotransmission. Before fusion, the vesicles dock to the
plasma membrane and proceed to the priming state, at which point the vesicles are
ready for fusion. The fusion occurs when the calcium channels are opened upon
arrival of an action potential.
1.2
The role of SNARE proteins in exocytosis
Neuronal exocytosis is mediated and regulated by a series of proteins, among
which SNARE (soluble N-ethylmaleimide-sensitive-factor attachment receptor) proteins play a central role. The formation of a ternary complex of SNARE proteins is
2
crucial for the overall process and the SNARE complex itself is the minimal machinery
of neuronal membrane fusion [17, 20, 62, 102, 138]. In neuronal exocytosis, the SNARE
complex consists of synaptobrevin 2 (Sb2), syntaxin 1A (Sx1A) and synaptosomalassociated protein 25 (SNAP25).
Sb2 is a vesicular SNARE protein [119], whereas Sx1A and SNAP25 are located in
the plasma membrane and form a complex [34, 38]. The core of a SNARE complex is a
bundle of four α-helices, each of them being a so-called SNARE motive [114]. Sb2 and
Sx1A each provide one motif and SNAP25 contributes two (Figure 1.2).
The SNARE complex has a highly conserved layer structure; each of the layers consists of four amino acid side chains which interact with each other. Due
to the composition of the layer structure, SNARE proteins are also classified as QSNAREs (Sx (Qa) and SNAP25 (Qb and Qc)) and R-SNARE (Sb, see Figure 1.3) [36].
The SNARE assembly proceeds from the cytosolic N-terminus to the membrane anchored C-terminus (zippering hypothesis) [48, 91].
Figure 1.2: Four-helix-bundle of the core SNARE complex [114]. Sb: Synaptobrevin (blue).
Sx: Syntaxin (red). Sn1: SNAP-25(N) (green). Sn2: SNAP-25(C) (green).
At the stage of the docked intermediate, SNARE proteins form the trans-complex,
in which the C-terminal transmembrane domains of the proteins are in separate membranes. During fusion, the trans-complex converts into the cis-complex form, in which
both transmembrane domains of Sx1A and Sb2 come together in the same membrane, and stimulates membrane merging [63]. The cis-complex is inactive for the
further membrane fusion and can be disassembled by AAA+ ATPase protein NSF (Nethylmaleimide-sensitive factor) [77] and its cofactor SNAP proteins (soluble NSF at3
Figure 1.3: Layer structure of the ternary SNARE complex [36]. Synaptobrevin (blue), Syntaxin (red) and SNAP-25 (green) assemble a helix-bundle with a conserved layer
structures, in which the +8 layer is the last layer before the C-terminal linker region
and the +7 layer is the first layer at the N-termini.
tachment proteins) [22] under ATP hydrolysis.
The SNARE disassembly will be discussed later in Section 1.4. The disassembled Sx and SNAP25 furthermore form the acceptor complex upon invocation by
SM (Sec1/Munc-18) family proteins and are available for the further vesicle docking
and fusion (Figure 1.4) [63]. SM proteins are not included in this study.
4
REVIEWS
Binding proteins?
ADP
ATP
Free SNARE clusters
Disassembly complex
α-SNAP
NSF
SM
proteins
Qa-SNARE
Qb-SNARE
Qc-SNARE
R-SNARE
Acceptor complexes
NSF
Cis-SNARE complexes
α-SNAP
Late regulatory proteins
(for example, complexins and synaptotagmin)
Vesicle
Loose trans-SNARE complexes
Acceptor membrane
Figure 1.4:
Tight trans-SNARE complexes
Trans → cis
(fusion)
Figure 3 | The SNARE conformational cycle during vesicle docking and fusion. As an example, we consider three
Q-SNAREs (Q-soluble N-ethylmaleimide-sensitive factor attachment protein receptors) on an acceptor membrane and
an R-SNARE on a vesicle. Q-SNAREs, which are organized in clusters (top left), assemble into acceptor complexes, and this
assembly process
might require SM (Sec1/Munc18-related)
proteins. Acceptor
complexes
interact
with the
vesicular
The SNARE
conformational
cycle during vesicle
docking
and
fusion
[63].
Three
R-SNAREs through the N-terminal end of the SNARE motifs, and this nucleates the formation of a four-helical transcomplex. Trans-complexes proceed from a loose state (in which only the N-terminal portion of the SNARE motifs are
Q-SNAREs
and(inQc)
in the
plasma
membrane
and by
one
R-SNARE
‘zipped (Qa,
up’) to aQb
tight state
whichlocalized
the zippering process
is mostly
completed),
and this is followed
the opening
of the
fusion pore. In regulated exocytosis, these transition states are controlled by late regulatory proteins that include
complexins
proteins that
bind to the surface
complexes)
and synaptotagmin
(which isare
activated
by an to
localized
in the(small
synaptic
vesicles
formofaSNARE
fusion
machinery.
Vesicles
docked
influx of calcium). During fusion, the strained trans-complex relaxes into a cis-configuration. Cis-complexes are
disassembled
by the AAA+by
(ATPases
associated
with various cellular
activities) protein
NSF (N-ethylmaleimide-sensitive
the plasma
membrane
SNARE
interaction,
forming
a loose
trans-SNARE comfactor) together with SNAPs (soluble NSF attachment proteins) that function as cofactors. The R- and Q-SNAREs are then
separated by sorting (for example, by endocytosis).
plex. At the priming stage associated with other late regulatory proteins, SNAREs
assemble
to a than
tight
are prepared
for fusion.
Upon
of an
is greater
100.trans-complex
Most of the evidenceand
is confined
Acceptor complexes:
intermediates
in thearrival
fusion pathway?
to qualitative approaches such as pull-downs and co-
How does the assembly of four unstructured SNAREs
false positives with ‘sticky’ proteins. Quantitative and/or
exocytic S. cerevisiae and neuronal SNARE complexes
action potential,
SNAREs
assemble
to theinto
cis-conformation
and promote
memimmunoprecipitations,
whichfully
are notorious
for yielding
a SNARE complex proceed?
Detailed studies
on
structural The
data about
these presumed SNARE–targetvitro have
that assembly
through a
brane fusion.
cis-SNARE-complex
is not in
active
forshown
fusion
and isproceeds
disassembled
protein complexes are therefore largely missing. To vali- defined and partially helical Qabc intermediate55–57, the
formation
of which
is rate limiting.
Although it is notThe
yet
date ATPase
these interactions,
detailed
structural
by AAA+
protein
NSF
withinformation
its cofactor
α-SNAP
under
ATP hydrolysis.
and a rigorous assessment of their in vivo relevance are known whether other SNAREs form such intermediate
acceptor
complexes,acceptor
it is probablecomplex
that they represent
a
required. Q-SNAREs are transformed into
disassembled
the active
by SM
Plasma-membrane SNAREs are not uniformly distrib- key step in the fusion pathway of all SNAREs — that
it is likely
that assembly
is an ordered,
sequential
uted
in the membrane, but are clustered proteins)
in nanodomains,
proteins
(Sec1/Munc18-related
andis,can
further
take part
in membrane
the stability of which depends on cholesterol51–53 . The reaction rather than a random collision of four differAdaptor protein-1 (AP1)
association of SNARE transmembrane ent SNARE motifs. Only when an acceptor scaffold is
fusion. homomeric
complex
domains has been reported, and this might contribute available in which the N-terminal ends of the SNARE
The AP1 complex, which is
to cluster formation54 . Secretory vesicles selectively dock motifs are structured is the final SNARE able to bind
one of four structurally
and fuse at such clusters51. It remains to be seen whether with biologically relevant kinetics and nucleate the
related protein complexes,
forms a bridge between the
cluster formation is a hallmark of all SNAREs or is a zippering reaction.
clathrin coat and membrane
Acceptor complexes are highly reactive and are therespeciality of plasma membranes and other membranes
components (cargo) during
that are rich in steroid lipids. Clustering achieves high fore difficult to characterize. For example, in vitro, the
the formation of clathrinlocal SNARE concentrations that might result in more neuronal acceptor complex readily recruits a second
coated vesicles at the transQa-SNARE, which results in a ‘dead-end’ Qaabc complex.
Golgi network.
efficient fusion.
NATURE REVIEWS | MOLECULAR CELL BIOLOGY
5
© 2006 Nature Publishing Group
VOLU M E 7 | SEP T E M BER 20 0 6 | 635
1.3
Calcium sensor protein: synaptotagmin-1
Neuronal exocytosis is triggered by the influx of calcium: Synaptic vesicles are
docked at the plasma membrane but do not undergo fusion until an action potential causes a transient increase in the intracellular calcium concentration. The calcium influx is sensed by synaptotagmin-1 (Syt1, 65 kDa) [19, 37, 79], which is linked
to the synaptic vesicle by their transmembrane domains (aa 58-79) in about seven
copies [119]. Syt1 possesses two calcium binding C2 domains, referred to as C2A (aa
140-265) and C2B (aa 271-421), which are connected to the transmembrane domain
by a long (61 residue) flexible linker [37]. C2A binds three and C2B two calcium
ions (Figure 1.5) [37, 95]. Syt1 binds to the acidic phospholipids of the membrane
in the presence of calcium [3, 13, 42, 56, 59, 73, 95]. C2B contains a polybasic lysine
patch (aa 324-327) proximal to the calcium binding site and has a specific affinity to
phosphatidylinositol-4,5-bisphosphate (PiP2) in a calcium independent manner (Figure 1.5) [6, 73, 104, 134, 141]. The polybasic stretch also binds phosphatidylserine (PS).
Moreover, Syt1 interacts with syntaxin and SNARE-complex-containing syntaxin in
the absence of calcium [6, 19, 21, 69, 125], albeit the binding is influenced to a limited
degree by calcium [19, 101].
Syt1 is well known as capable to increase the rate of exocytosis by several orders
of magnitude [96], but the molecular mechanism is still unclear. Two different priming
models are presently being discussed. In the first model, a partial SNARE assembly is
formed and arrested at this state, which is assumed to be caused by either an energy
barrier or interaction with complexin or synaptotagmin. After receiving calcium, Syt1
may (i) bind to the SNARE complex and the plasma membrane (PiP2, PS), displace
the inhibitory complexin and promote the full SNARE zippering [111, 134, 135]; or (ii)
bind to the plasma membrane, induce curvature stress close to the membrane contact
area and lower the energy barrier [61, 78, 82]; (iii) alternatively cross-link the vesicle
membrane and the plasma membrane and promote fusion by compensating the membrane charge [3]. The second model proposes tethering/docking of the vesicle to the
plasma membrane by Syt1 trans-binding to PiP2 contained in the plasma membrane
6
without SNARE assembly. Calcium activates both C2 domains and thus cross-linking
of the membrane to further promote the SNARE assembly. Hereby, Syt1 functions as
distance regulator [111, 141].
To dissect the molecular mechanism of Syt1, SNARE-mediated fusion was performed using reconstituted liposomes. Contradictory effects of Syt1 have been reported. Several studies have shown that tethering/docking of Syt1 to the plasma
membrane promotes SNARE assembly [61, 78]. However, tethering was not observed
as a separate intermediate because of the presence of the SNARE proteins. Inhibition of fusion has also been observed in several fusion experiments with SNAREs and
Syt1 [67, 111]. This led to further complications in the study, because Syt1 may also
bind in cis to the vesicular membrane and cis-binding competes with trans-binding.
Since trans-binding of Syt1 is required by all of the different fusion models, a balance
between cis- and trans-binding may play an essential role in neurotransmission. Therefore, this study will focus on the cis- and trans-membrane interaction of Syt1.
7
PI(4,5)P2 Increases the Ca2ⴙ Affinity of Synaptotagmin
Downloaded from www.jbc.org at Max Planck Inst.Biophysikalische Chemie,Otto Hahn Bibl,Pf.2841,37018 Goettingen, on August 7, 2012
Figure 1.5: 3D structure of synaptotagmin-1. [95] Synaptotagmin-1 possesses one transmembrane domain and two C2 domains, called C2A and C2B. All the three domains
are linked subsequently to each other by two flexible linkers. The C2A domain
binds three calcium ions, while the C2B domain binds two calcium ions. Mutant (K326,327A) of the calcium independent binding site of C2B is shown as
“KAKA” (blue). S342C shows the position of the cysteine mutant, at which the
fluorescence dye can be attached (green).
25750 JOURNAL OF BIOLOGICAL CHEMISTRY
VOLUME 284 • NUMBER 38 • SEPTEMBER 18, 2009
8
1.4
NSF and α-SNAP
NSF (N-ethylmaleimide-sensitive factor) belongs to the AAA+ ATPase family and
its function in neuronal exocytosis is to accomplish disassembly of the SNARE complex
under ATP hydrolysis [81, 115]. NSF consists of two AAA+ domains, referred to as
D1 domain (aa 206-477) and C-terminus D2 domains (aa 478-744), and one N-terminus
domain (aa 1-205) [118, 137]. For disassembly, NSF forms a homohexamer [39, 40, 49]
and provides the energy for disassembly by its D1 domain, while the C-terminus D2
domain is mainly responsible for the hexamerization [84].
NSF does not bind to the SNARE complex directly. To transfer the energy from
the ATP hydrolysis to the SNARE complex, NSF must interact with its cofactor αSNAP (soluble NSF attachment proteins) [22], which consists of fourteen subsequent
α-helices [80, 132] and connects the SNARE complex and the NSF hexamer with two
more copies of itself [131]. Altogether, NSF hexamer, three α-SNAPs and one SNARE
complex form an complex, called 20S complex or 20S particle [40, 49, 57, 131]. The
N-termini of the NSF hexamer attach to the α-SNAPs at their C-termini [8, 84, 136]
and energy is transferred to the SNARE complex via interaction of the N-terminus
of α-SNAP with the C-terminal side of SNARE complex near the membrane [52]. To
process the SNARE disassembly, α-SNAP has to attach to the membrane through the
hydrophobic loop located at its N-terminus (Figure 1.6) [132]. The disassembly is promoted by the conformation change of the NSF hexamer concerted with the ATP hydrolysis. This movement is transferred to the SNARE complex and the membrane by
α-SNAPs [49, 115].
Disassembly of the SNARE complex increases the number of free SNARE proteins,
and thus plays a positive role for vesicle priming [18, 66, 120, 133]. On the other hand,
free α-SNAP has been shown to inhibit the vesicle fusion in Drosophila [4] and the
exocytosis of dense-core vesicles in PC12 cells [9]. However, how this inhibition of
membrane fusion occurs is remaining unclear.
9
␣-SNAP Contains a Membrane Attachment Site
FIGURE 4. A hydrophobic loop in the N-terminal region of SNAP might serve as membrane attachment site. a, structure-based sequence alignment of the
N-terminal portion of SNAP proteins from different organisms indicates that the hydrophobic loop between the first two helices is conserved. At the top, boxes
indicate the first two helices and the connecting loop from the Sec17 crystal structure. The yeast Sec17 (SaCe_Sec17, Saccharomyces cerevisiae, gi 6319421) was
aligned1.6:
with several
fungal andattachment
animal SNAP homologs:
LoLe_Sec17,
Lodderomyces
gi 149246407;
Figure
Membrane
site of
α-SNAP
in theelongisporus,
presence
of the CaAl_Sec17,
SNARE Candida
com- albicans, gi 68475136;
CaGl_Sec17, Candida glabrata, gi 50287489; CiIn_␣-SNAP, Ciona intestinalis, gi 198424864; DrMe_␣-SNAP, Drosophila melanogaster, gi 17737681; TrAd_␣SNAP, Trichoplax adhaerens, gi 196014845; StPu_␣-SNAP, Strongylocentrotus purpuratus, gi 72082731; BrFl_␣-SNAP, Branchiostoma floridae, gi 219425561;
plexBos
[132].
The
yeast
Sec17
from Baker’s
asaromatic
fourteen
and BoTa_␤-SNAP,
taurus, gi
423236;
BoTa_
␣-SNAP,(α-SNAP)
Bos taurus, gi 423230.
Arrowheadsyeast
indicateis
thestructured
highly conserved
residues that were mutated
to serines. b, schematic drawing of the crystal structures of Sec17 from Baker⬘s yeast (PDB code 1QQE, S. cerevisiae) (56) and of the membrane-embedded
neuronal SNARE
complex including
its transmembrane
(Ref. 57, PDBof
codes
3HD7 and 3HD9). loop
This illustration
indicates
that the loop between the first
subsequent
α-helices
(α1-α14)regions
and consists
a hydrophobic
localized
between
two helices of Sec17 might touch the membrane when the protein is bound to the SNARE bundle. Note that the illustration is largely based on the model of the
␣-SNAP: SNARE complex interaction given in Marz et al. (20).
the α1- and α2-helices. The loop attaches to the membrane when α-SNAP binds the
conserved SNARE
phenylalanines
(residues
27 and 28) for2,the
polar
complex
(Synaptobrevin
Syntaxin
amino acid serine (␣-SNAPF27S,F28S).
disassembly
theLost
SNARE
complex by NSF.
Both Mutated
␣-SNAPsof
Have
the Lipid-mediated
Efficiency Boost—We found that both mutants were able to disassemble soluble SNARE complexes. ␣-SNAPF27S,F28S turned out
to promote disassembly of soluble SNARE complexes as efficiently as wild-type ␣-SNAP (Fig. 5a), whereas ␣-SNAPdel was
somewhat less efficient. The latter was not surprising, because a
deletion of residues 1–28 had been reported to hamper SNAP
efficiency before (25). We therefore chose to concentrate on the
less severe ␣-SNAP mutant, ␣-SNAPF27S,F28S. We directly
compared its disassembly kinetics in solution and on liposomes
of ␣-SNAPF27S,F28S and of ␣-SNAPwt, using the fluorescence
anisotropy set-up. Knowing that ␣-SNAPwt efficiency increases
20-fold on liposomes, we employed ⬃20-fold less of the respective ␣-SNAP (60 nM) for the experiments on liposomes, leaving
everything else as in solution. Remarkably, whereas 60 nM
31822 JOURNAL OF BIOLOGICAL CHEMISTRY
10
wt
␣
-SNAP
efficiently promoted
disassembly
1a
and SNAP-25),
and facilitates
the of membrane-inserted SNARE complexes, the same amount of ␣-SNAPF27S,F28S
did not support disassembly on liposomes at all (Fig. 5b).
Increasing the concentration of ␣-SNAPF27S,F28S led to a gradual increase of disassembly speed on liposomes. When we also
tested the ␣-SNAPdel mutant, notwithstanding its reduced
overall ␣-SNAP efficiency, we found that the membrane-mediated ␣-SNAP-potentiation was also abolished (supplemental
Fig. S3). Together, these findings suggest that the arm-like
structure at the N-terminal tip of ␣-SNAP might serve as a
hitherto unknown membrane attachment site.
The Membrane Boost Is Conserved for Other SNAP Proteins—
We next asked whether the membrane boost is conserved for
other homologs of the disassembly machinery. To answer
this question, we focused on the mammalian brain-specific
SNAP isoform ␤-SNAP and the SNAP homolog of yeast,
Sec17 (13, 45).
VOLUME 284 • NUMBER 46 • NOVEMBER 13, 2009
1.5
Fluorescence correlation spectroscopy
1.5.1
Fluorescence correlation spectroscopy (FCS)
Fluorescence correlation spectroscopy (FCS) is a technique with single molecule
level resolution, which is used in this study to temporally characterize the molecular
dynamics of biomolecules in vitro in terms of concentration, diffusion time, and interaction with other biomolecular components [5, 31, 76, 98].
The core of FCS is not the characterization of the individual fluorescence event
itself–as in time-correlated single photon counting (TCSPC)–but rather the temporal
fluctuation of the total fluorescence intensity I (t) [31, 74, 75] caused by rotational- or
translational diffusion [29,75], population of the triplet state [14,25,130], chemical reactions or interactions [46,47,50,93], Förster Resonance Energy Transfer (FRET) [121,129].
The fluctuation of a single molecule’s emission can be observed only when the sample
is strongly diluted, so that the diffusion of a single particle into (on) and out of (off) the
effective volume cannot be neglected. The ideal sample concentration for FCS measurements is from subnanomolar to submicromolar.
To analyze experimental data, the photons collected by the detector are added up
in time intervals of 5 µs each and a temporal fluorescence trace I (t) is constructed from
these intervals. The fluctuation can be described as [31, 50]:
δI (t) = I (t) − h I (t)i
(1.1)
where the h I (t)i is the mean fluorescence intensity I (t) over the entire time of the
experiment (0 – T). Autocorrelation is a characterization of data or functions in terms
of the “self-similarity”, in this case of the signal I at time t, I (t), compared to the signal
I at a time shift t + τ, I (t + τ ) (Figure 1.7). The autocorrelation is computed as [31, 50]:
G (τ ) =
h I (t) · I (t + τ )i
h I (t)i
2
−1 =
11
hδI (t) · δI (t + τ )i
h I (t)i2
(1.2)
I (t)
I(t+t0)
I(t+t1)
I(t+t2)
I(t+t3)
t
0
G(t)
x
0
tD
lag time (t)
Figure 1.7: Fluorescence autocorrelation-spectroscopy. The autocorrelation function of a fluorescence signal I (t) provides a quantitative measure of the similarity of this very
signal with an instance of itself I (t + τ ) shifted by a temporal offset τ. A schematic
fluorescence trace I (t) and its time-shifted traces I (t + τx ) (x = 0 - 3) are shown
above (left). The autocorrelation G (τ ) is calculated by multiplication of I (t) with
I (t + τ ) over the entire trace (purple line), averaging the results and normalizing
by the square of the average intensity. For τ = 0, G (0) =
1
N
and at half of the decay τ
= τD . The temporal offset τ, also referred to as the “correlation time”, is the abscissa
value and is usually represented in log scale (right).
12
In this study, the liposomes and the SVs diffuse in a three-dimensional Brownian
motion and the analytical expression of the autocorrelation function hence is [76]:
G (τ ) =
1
hC i · π 3/2 · w02 · z0
·
4Dτ
1+ 2
r0
! −1
·
4Dτ
1+ 2
z0
!−1/2
(1.3)
In this equation, hC i is the average of the particle concentration in the effective focal
volume Ve f f , while π 3/2 · w02 · z0 is Ve f f as calculated by the integration of the emission
distribution over the focal volume V. D represents the molecular diffusion coefficient
of the particle, and w0 and z0 the radius of the focal volume in lateral and axial directions, respectively. The focal volume is a property of the setup (laser beam and
objective).
The time-independent part of this function can be expressed as
1
N,
where N is the
average particle number in Ve f f and can be calculated as N = hC i · Ve f f . The diffusion
time τD , for which a particle stays in the focal volume, is inversely proportional to the
diffusion coefficient D [30]:
τD =
w0 2
4D
τD,z =
z0 2
4D
(1.4)
τD is the diffusion time in the lateral direction and τD,z in the axial direction. With these
diffusion times, Equation 1.3 can be reformed as [50]:
1
G (τ ) =
·
N
The ratio of
w0
z0
τ
1+
τD
−1/2
−1 τ w0 2
· 1+
·
τD z0 2
(1.5)
is a property of the experimental setup and was determined directly
using immobilized gold beads (∅ = 20 nm) on a micro-XYZ-stage for this study by
Dr. W. H. Pohl [92]. The reflection of the laser beam by the gold beads can be observed
only within the focal volume. The value of
w0
z0
for the water immersion objective used
in this study (UPlanSApo 60x/1.2w, Olympus) is 0.25 [92].
The time dependent part of the correlation function approaches unity at τ = 0, and
thus the ordinate value gives the reciprocal of the particle number within the effective
volume:
G (0) =
13
1
N
(1.6)
At τ = τD , the time dependent part of the autocorrelation function is about 0.5,
neglecting the reciprocal square root, so that the diffusion time τD can be read off at
the half-value of the decay. Figure 1.7 shows a schematic of the autocorrelation curve.
1.5.2
Fluorescence cross-correlation spectroscopy (FCCS)
In case of two-color experiments, two different dyes are used for labeling the sample particles, either in two different groups of sample components or at different positions of the same sample particles. The two dyes are excited simultaneously either with
two (one-photon excitation, OPE) lasers or with a single (two-photon excitation, TPE)
laser. Using fluorescence cross-correlation spectroscopy (FCCS), the number of interacting particles Nx which accordingly emit both colors from spatial proximity after, for
instance, fusion or tethering, can be determined. It should be noted at this point, that
the interaction of the spatially close emitters might impact on the characteristics of the
emission (fluorescence lifetime and intensity). However, for the immediate treatment
of FCCS in this subsection, these effects shall not be considered. The FCCS function is
analogous to the FCS function, so that a similarity comparison is performed between
the separated fluorescence traces of two colors [50, 107]:
GRG (τ ) =
hδIR (t) · δIG (t + τ )i
h IR (t)i · h IG (t)i
(1.7)
GRG (τ ) is the amplitude of the cross-correlation. IR (t) and IG (t) are the intensity functions of the separate red and green channels (Figure 1.8).
Equation 1.5 is fitted to the cross-correlation data and, from the resulting GRG (0),
the multicolor particle number Nx can be calculated as [99]:
GRG (0) =
Nx
( NR + Nx ) · ( NG + Nx )
(1.8)
NR and NG are the numbers of the free red and green emitters, respectively.
The sum of Ni (i = R, G) and Nx can be calculated from the autocorrelation of the
14
respective channel:
1
GR (0)
1
NG + Nx = NG (0) =
GG ( 0 )
NR + Nx = NR (0) =
(1.9)
(1.10)
Nx can subsequently be obtained as:
Nx = GRG (0) · NR (0) · NG (0)
IR(t)
IG(t+t0)
IG(t+t1)
IG(t+t2)
IG(t+t3)
t
0
G(t)
x
(1.11)
0
lag time (t)
Figure 1.8: Fluorescence cross-correlation-spectroscopy. The FCCS concept is analogous to
FCS but with two separate fluorescence channels (left).
From the results of
FCCS (blue solid line) and of FCS for each color (red and green dash lines), the
number of the particles with both red and green dyes Nx can be calculated (right).
1.5.3
Two-photon excitation (TPE)
FCS measurements require a low sample concentration (nM) and a small detection
volume, so that a fluctuation of the individual dye’s emission can be observed clearly.
15
The latter can be achieved using an objective with high numerical aperture (NA,
NA = 1.2 in this study) in a confocal microscope [100], minimizing the lateral extent
of the excitation beam to several 102 nm. Furthermore, a pinhole is needed to limit the
axial dimension of the focal volume. Alternatively, two-photon excitation (TPE), due
to its excitation property, can be used for FCS to reduce the excitation volume and thus
the effective volume of the focus [10, 23, 27, 55].
During two-photon excitation, the fluorescence dye absorbs, in one excitation process, two photons of theoretically double the wavelength used for one-photon excitation within a sub-femtosecond time range (Figure 1.9a) [44]. To achieve this, a pulsed
laser is required for drastically increasing the photon density in the focus area. The
excitation probability is proportional to the square of the laser’s intensity, so that only
the middle part of the focal volume can be excited. The distribution of the emission
light depends on the laser beam and the objective, and can be approximated as a 3D√
Gaussian. Thus the effective excitation volume is decreased by a factor of 2 in each
direction for TPE:
OPE:
TPE:
Ve f f = π 3/2 · w02 · z0 (Page 13, Subsection 1.5.1)
π 3/2
Ve f f =
· w02 · z0
2
(1.12)
(1.13)
and the diffusion time is accordingly shorter by a factor of two in each direction (Figure 1.9b):
OPE:
TPE:
w0 2
4D
w0 2
τD =
8D
τD =
z0 2
4D
z 2
= 0
8D
τD,z =
τD,z
(Page 13, Equation 1.4)
(1.14)
(1.15)
The typical VeOPE
f f for OPE is about 1 fl.
For our two-photon confocal microscope, the lateral radius w0 was determined
using the fluorescence dye Rhodamine Green (RG) with a known diffusion coefficient
DRG = 2.8 × 10−6 cm2 s−1 [24, 100] and calculated using equation 1.15. The VeTPE
f f was
calculated with equation 1.13 and the value is about 0.3 fl, which perfectly confirmed
√ √
the expected decrease in detection volume by a factor of 8 ( 2 in each direction)
compared to OPE.
16
a
OPE TPE
Em
b
S1
z0
w0
S0
Figure 1.9: OPE compared with TPE in terms of the excitation module and the effective volume. a. Jablonski-diagram for OPE and TPE. b. Schematic of the effective excitation
volume by OPE (green) and TPE (red).
Two-photon excitation allows simultaneous excitation of two fluorescence dyes
with one laser beam and protects the biological sample against denaturation throughout most of the path of the laser beam due to its low photon energy and its small
excitation volume.
1.5.4
Förster resonance energy transfer (FRET)
Förster resonance energy transfer (FRET) is a non-radiative energy transfer between a donor fluorophore and an acceptor fluorophore. The donor is excited to its
electronically excited state, transfers the energy to the acceptor nearby and thus excites the acceptor [41]. In this process, the donor fluorescence is quenched and only
the acceptor is able to fluoresce (Figure 1.10). For FRET to occur, the spectra of the
donor emission and the acceptor absorption need to overlap, the distance between the
donor and the acceptor has to be relatively short (10-100 Å), and relative dipole–dipole
orientations have to be favorable for coupling [122].
In general, the experimentally observable rate constant k exp of the decay of the
17
FRET
S1
S1
×
S0
S0
Acceptor
Donor
Figure 1.10: Jablonski-diagram of FRET donor and acceptor. As the excited state’s energy
is transferred from the donor to the acceptor, donor emission is quenched and
acceptor fluorescence enhanced.
donor’s (D) and acceptor’s (A) respective S1 state populations is a sum of the fluorescence (Fl) rate constant and other rate constants, for instance internal conversion (IC),
intersystem crossing (ISC):
D
D
D
k exp
= kD
Fl + k IC + k ISC
(1.16)
A
A
A
k exp
= k Fl
+ kA
IC + k ISC
(1.17)
If FRET occurs, the depletion of the donor’s electronically excited state is accelerated:
0
D
D
D
k exp
= kD
Fl + k IC + k ISC + k FRET
(1.18)
The fluorescence quantum efficiency can be calculated as k Fl /k exp and the observed fluorescence lifetime τexp is the inversion of the emission rate constant k exp .
Thus, the donor emission count rate and lifetime τD decrease under FRET.
The FRET distance is the donor-acceptor-distance at which the FRET quantum efficiency k FRET /k exp is 50% [11]. FRET is, of course, most pronounced below this distance.
In the Syt1 tethering experiment, the liposomes were labeled with either 1% TR-DHPE
or 1.5% OG-DHPE in such a way that upon fusion the distance between TR and OG is
about 7 nm. According to a previous study using TR-OG FRET pair [23, 55], FRET can
be observed upon fusion. Figure 1.11 shows the excitation and emission spectra of TR
18
Normalized Intensity
and OG.
100
Oregon Green Excitation
Oregon Green Emission
Texas Red Excitation
Texas Red Emmission
50
0
300
350
400
450
500
550
600
650
700
750
λ/nm
Figure 1.11: Excitation and emission spectra of TR-DHPE and OG-DHPE.
1.6
Outline of this Study
This study focuses on the functional mechanism of Syt1 with the particular em-
phasis on the cis- and trans-membrane interaction of Syt1. To achieve this, molecular
requirements for the calcium dependent and independent binding properties are investigated using reconstituted liposomes with a sensitive tethering assay based on two
photon-fluorescence cross correlation spectroscopy (TP-FCCS). Tethering is measured
between the liposomes bearing Syt1 and the liposomes containing acidic phospholipids, in this case phosphatidylserine (PS) and phosphatidylinositol-4,5-bisphosphate
(PiP2), under the influence of calcium. The host liposomes of Syt1 contain either no
acidic phospholipids or 20% PS to obtain either trans- or cis-binding. Furthermore,
interaction between membrane anchored Syt1 and membrane anchored SNARE proteins (Syntaxin 1A (Sx1A), a 2:1 complex consisting of two Sx1A and one SNAP-25,
and a ternary complex consisting of Sx1A, SNAP-25 and Synaptobrevin 2 (Sb2) lacking transmembrane domain) is tested. To investigate the binding function of different
binding sites, mutants at C2A and C2B domains of the full-length Syt1 are used instead
of the wild-type Syt1. In addition, cross-linking mediated by soluble Syt1 (aa 1-97) is
measured to complete the study by comparison with the previous publications.
19
Associated with the Syt1 project, two preliminary measurements are performed in
this study, aimed at characterizing the mouse synaptic vesicle (SVs) by average protein mass, and monitoring the chromaffin granules (CGs) docking with the large liposomes mediated by partial SNARE assembly arrested by α-SNAP. Purified mouse
SVs are labeled with FM 1-43 fluorescence dye and measured with fluorescence correlation spectroscopy (FCS) to determine the vesicle number. Combining the protein
concentration of the same sample, an average vesicular mass of proteins can be calculated. This experiment introduces a nice comparison with the previous study of rat
SVs and establishes a vesicular base for the further characterization of the composition
of proteins as well as lipids in a single vesicle manner.
Using FCCS tethering assay, a stable docked state between CGs and large liposomes can be monitored for the first time. α-SNAP is reported to be able to inhibit CG
fusion by the lipid mixing assay. Here, docking can be observed under the same conditions. This observation suggests a partial assembly model of the SNARE complex
mediated by α-SNAP, which inhibits CG fusion, but not CG docking.
20
Chapter 2
Materials and Methods
2.1
Materials
2.1.1
Chemicals
All the standard chemicals used in this study were purchased from the following companies: Sigma (Deisenhofen, Germany), Sigma-Aldrich (St. Louis, USA),
AppliChem (Darmstadt, Germany), Merck (Darmstadt, Germany), Roth (Karlsruhe,
Germany), Biorad (Richmond, USA), Serva (Heidelberg, Germany), Boehringer (Ingelheim, Germany ), Fluka (Switzerland) and Anatrace (USA). The column material
Sephadex G50 was purchased from GE Healthcare (Freiburg, Germany) and the molecular weight protein standards (SM0671) was purchased from MBI Fermentas (St. LeonRot, Germany).
2.1.2
Phospholipids
All the phospholipids used in the tethering experiments were purchased from
Avanti Polar Lipids Inc. (Alabama, USA), except Texas Red phosphatidylethanolamine
and Oregon Green phosphatidylethanolamine, which were purchased from Invitrogen
21
CHAPTER 2. MATERIALS AND METHODS
Molecular Probes (see Table 2.1).
Abbreviation
Phospholipid
Company
PC
L-α-Phosphatidylcholine
Avanti Polar Lipids
PE
L-α-Phosphatidylethanolamine
Avanti Polar Lipids
PS
L-α-Phosphatidylserine
Avanti Polar Lipids
PI
L-α-Phosphatidylinositol
Avanti Polar Lipids
PiP2
Phosphatidylinositol-4,5-
Avanti Polar Lipids
bisphosphate
TRPE
Texas Red-PE
Invitrogen Molecular Probes
OGPE
Oregon Green-PE
Invitrogen Molecular Probes
NBD-DOPE
1,2-dioleoyl-sn-glycero-3-
Avanti Polar Lipids
phosphoethanolamine-N-(7nitrobenz-2-oxa-1,3-diazol-4-yl)
Rhodamine-DOPE
1,2-dioleoyl-sn-glycero-3-
Avanti Polar Lipids
phosphoethanolamine-Nlissamine rhodamine B sulfonyl
ammonium salt
Table 2.1: Phospholipids used for reconstitution of liposomes.
2.1.3
Proteins
All the protein constructs used in this study were from Rattus norvegicus. The proteins were cloned in the pET28a (Novagen) vector and expressed in Escherichia coli
strain BL21 (DE3) except the ∆N complex, in which syntaxin 1A (Sx1A) (183-288) and
synaptobrevin 2 (49-96) were cloned in pET Duet-1 (Novagen), and SNAP 25A was
cloned in pET28a.
Expression constructs of the full-length synaptotagmin-1 (Syt1) (1–421) and of
the soluble domain of Syt1 (97–421), have been described elsewhere [111]. Also the
22
following calcium mutants of the full-length Syt1 have been described earlier [111]:
a*B (D178, 230, 232A), Ab* (D309, 363, 365A), a*b* (D178, 230, 232, 309, 363, 365A), and
KAKA mutant (K326, 327A). The constructs for the neuronal SNAREs were the SNARE
motif of syntaxin 1A with its transmembrane domain (183–288), a cysteine-free variant
of SNAP-25A (1–206), and synaptobrevin 2 without its transmembrane domain (1–96).
The Syt1 (97–421) single cysteine variant (S342C) was obtained after first removing the
single native cysteine (C278S) and then introducing a point mutation at position 342.
All proteins were expressed in Escherichia coli strain BL21 (DE3) (Novagen) and
purified using Ni2+ -nitrilotriacetic acid beads (Ni-NTA, GE Healthcare) followed by
ion exchange chromatography (IEXC) on the Aekta system (GE Healthcare) as described in the literature [91, 111], with a few modifications. The Syt1 single cysteine
variant (97–421, S342C) was further labeled with Alexa Fluor 488 C5 maleimide. This
was done by first dialyzing the proteins against the AF-labeling buffer (page 28, Table 2.3). The dialyzed protein solution was then incubated with the fluorophore for
2 h at RT and separated from the free dye using a Sephadex G50 superfine column.
The labeling efficiency was about 40% [111, 140, 141]. Sx1A (183–288) and synaptobrevin 2 (1–96) were purified by IEXC in the presence of 15 mM CHAPS. The binary
complex (2:1) containing Sx1A (183–288) and SNAP-25A (1-206) was assembled from
purified monomers and subsequently purified by IEXC in the presence of 1% CHAPS.
The ∆N complex was co-expressed with pET Duet-1 and pET28a vectors and purified by IEXC with 50 mM n-octyl-β-D-glucoside [91]. The ternary SNARE complex
consisting of Sx1A (183–288), SNAP-25A and synaptobrevin 2 (1–96) was generated by
incubating of the binary complex and synaptobrevin 2 (1–96) in a ratio of 1:2 over night
at 4°C. The excess synaptobrevin 2 was removed with Sephadex G50 superfine column
during liposome reconstitution.
All the SNARE proteins including different SNARE complexes, and α-SNAP and
its mutants were provided by Dr. Alexander Stein and Dr. Yongsoo Park (neurobiology
department) [33, 35, 91, 105, 111, 128, 132]. Full-length Syt1 (Syt1/1-421/PET28a/NdeIXhoI/His) was cloned by the neurobiology department and expressed in this study.
23
The expression and purification were described in page 30, subsection 2.2.1. The Syt1
purification protocol was modified from the publications and the dissertation of Dr.
Alexander Stein [95, 111]. The mutants of the full-length Syt1 and the soluble Syt1
were provided by Dr. Geert van den Bogaart [111, 140, 141]. The sequences of all the
proteins are given in Table 2.2.
Abbreviation
Protein
Sequence
Sb1−96
Synaptobrevin 2 1-96
1-96
[35]
SNAP25A
SNAP 25A
1-206, C84, 85, 90, 92S
[33]
Sx1A
Syntaxin 1A
183-288
[105]
Sx1A
183-288
SNAP25A
1-206, C84, 85, 90, 92S
Sb2 49-96
49-96
2x Sx1A
183-288
1x SNAP25A
1-206, C84, 85, 90, 92S
Syt1
Synaptotagmin-1
1-421
a*B
Synaptotagmin-1 a*B
1-421, D178, 230, 232A
Ab*
Synaptotagmin-1 Ab*
1-421, D309, 363, 365A
a*b*
Synaptotagmin-1 a*b*
1-421, D178, 230, 232,
∆N complex
2:1 complex
[91]
[111, 128]
this study
309, 363, 365A
KAKA
Synaptotagmin-1 KAKA
1-421, K326, 327A
C2AB
Soluble synaptotagmin-1
97-421
AF-C2AB
Soluble synaptotagmin-1
97-421,
[111, 140, 141]
S342C-Alexa
Fluor 488
α-SNAP
α-SNAP
1-295
α-SNAP33−295
α-SNAP deletion
33-295
α-SNAPF27,28S
α-SNAP mutation
1-295, F27, 28S
Table 2.2: Proteins used in this study.
24
[132]
2.1.4
Buffers
The buffers used throughout this study are listed in Table 2.3.
Table 2.3: Buffers used in this study.
Buffer
Concentration
Components
Reconstitution of liposomes [105, 138]
Chloroform/MeOH
DTT stock solution
HP150
HPCholate5
2/3 (v/v)
Chloroform
1/3 (v/v)
MeOH
1 mM
20 mM
HEPES
150 mM
KCl
2 mM
DTT
20 mM
HEPES
150 mM
KCl
2 mM
DTT
5% (w/v)
HPCholate1.5
DTT in ddH2 O (Milli-Q)
20 mM
Sodium Cholate
HEPES
150 mM
KCl
2 mM
DTT
1.5% (w/v)
Sodium Cholate
Tethering experiments
EGTA buffer
20 mM
HEPES
150 mM
KCl
2 mM
DTT
1 mM
EGTA
continued on next page
25
Buffer
Concentration
Calcium buffer
20 mM
CG buffer
Components
HEPES
150 mM
KCl
2 mM
DTT
1 mM
EGTA
1.1 mM
CaCl2
20 mM
HEPES, pH 7.4
120 mM
20 mM
5 mM
potassium glutamate
potassium acetate
MgCl2
Purification of synaptotagmin-1 (1-421)
PMSF stock solution
Extraction buffer
200 mM
20 mM
500 mM
Wash buffer
NaCl
Imidazole
20 mM
Tris, pH 7.4
20 mM
1% (w/v)
Thrombin stock solution
Tris, pH 7.4
20 mM
500 mM
Elution buffer
PMSF in EtOH
20 mM
NaCl
Imidazole
CHAPS
Tris, pH 7.4
500 mM
NaCl
400 mM
Imidazole
1% (w/v)
CHAPS
5 mg/ml
Trombin in 50% (v/v) Glycerol
26
Buffer
Desalting buffer
Concentration
20 mM
300 mM
1% (w/v)
Components
Tris, pH 7.4
NaCl
CHAPS
1 mM
EDTA
1 mM
DTT
Ion Exchange Chromatography (IEXC)
Buffer A
20 mM
1 mM
1% (w/v)
Buffer B
20 mM
1 mM
1% (w/v)
1M
Tris, pH 7.4
DTT
CHAPS
Tris, pH 7.4
DTT
CHAPS
NaCl
Schägger-gel for SDS-PAGE [108]
Sample buffer
50 mM
4% (w/v)
0.01% (w/v)
12% (v/v)
2% (v/v)
Gel buffer
3M
0.3% (w/v)
Tris, pH 6.8
SDS
Serva Blue G
Glycerol
β-Mercaptoethanol
Tris, pH 8.45
SDS
Anode buffer
2M
Tris, pH 8.9
Cathode buffer
1M
Tris
1M
Tricine
1% (w/v)
SDS
27
Buffer
Concentration
Coomassie staining solution
Coomassie destaining solution
0.2 % (w/v)
Components
Coomassie Brilliant Blue R
25% (v/v)
EtOH
10% (v/v)
Acetic acid
65% (v/v)
ddH2 O
20% (v/v)
EtOH
5% (v/v)
Acetic acid
1% (v/v)
Glycerol
SV purification
HB-100
25 mM
HEPES, pH 7.4
100 mM
KCl
1 mM
DTT
Lowry-Peterson protein determination [90]
Lowry solution I
189 mM
68 mM
8 mM
1% (w/v)
hline Lowry solution II
hline Lowry solution III
250 mM
100 ml
1 ml
Lowry solution IV
2N
1:1
Na2 CO3
NaOH
Na2 -Tartrate · 2 H2 O
SDS
CuSO4 · 5 H2 O
Lowry solution I
Lowry solution II
Folin-Ciocalteu’s phenol reagent
diluted with ddH2 O
Labeling buffers
FM 1-43 stock solution
AF-labeling buffer
1 mM
50 mM
FM 1-43 in ddH2 O
HEPES, pH 7.4
500 mM
NaCl
100 µM
Tris(2-carboxyethyl)phosphine
28
2.1.5
Instruments, filters, columns and others
The most important instruments and the filters as well as the columns for protein
purification and for the tethering experiments are listed in Table 2.4.
Table 2.4: Instruments, filters, columns and miscellanea.
Item
Supplier
Instruments
Electrophoresis chamber Mini-Protean II
Biorad (Richmond, USA)
Power Pac 300
Biorad (Richmond, USA)
Nanodrop-Spectrometer 1000
Thermo Scientific (Germany)
Novaspec II photometer
Parmacia Biotech (Germany)
Cary 5E, UV-Vis-NIR
Varian (Germany)
Aekta system
GE Healthcare (Germany)
J6-MI Centrifuge
Beckman Coulter (Germany)
Sorvall RC-5B Refrigerated Superspeed Centrifuge
Sorvall SS34
Sorvall F14-6x250y
SW41T1 rotor
Beckman Coulter (Germany)
Components for FCS measurements
Chameleon Ti:Sa laser system
Coherent (California, USA)
Solid State Thermoelectric Thermal Control Unit
Coherent (California, USA)
T225P
IX71 inverted microscope
Olympus (Germany)
UPlanSApo 60x/1.2w water immersion objective
Olympus (Germany)
Avalanche photodiode (APD, SPCM-AQR-13)
Perkin-Elmer (Canada)
PRT 400, 4-channel router
PicoQuant GmbH (Germany)
TimeHarp200, TCSPC card
PicoQuant GmbH (Germany)
29
Item
Supplier
Filters
715 DSCPXR dichroic mirror
AHF (Germany)
590 DCXR dichroic mirror
AHF (Germany)
E700SP2 short pass filter
AHF (Germany)
Ultra-broadband dielectric mirror, 650-1130 nm
Newport (USA)
HQ 645/75 bandpass filter
AHF (Germany)
HQ 535/50 bandpass filter
AHF (Germany)
Columns
Sephacryl S-1000 Superfine HR
Econo-column, 0.5cm x 10cm
Biorad, (Richmond, USA)
Econo-column, 2.5cm x 10cm
Biorad, (Richmond, USA)
Miscellanea
Coverslip 18 x 18 mm
Menzel-Gläser (Germany)
Ni2+ -nitrilotriacetic acid beads (Ni-NTA)
2.1.6
Software
The software products used for data analysis were MATLAB 2009b (The MathWorks, Inc.), Origin 8.6G (MicroCal Inc.), Microsoft Office Suite 2010 (Microsoft Corp.)
and SigmaPlot (Systat Software Inc.).
2.2
Methods
2.2.1
Expression and purification of synaptotagmin-1
Synaptotagmin-1 (Syt1) was cloned in the pET28a vector (Novagen) and transformed via heat shock in the expression cell Escherichia coli strain BL21 (DE3, Novagen,
see page 22, subsection 2.1.3). Transformation was performed with the standard pro30
tocol of the neurobiology department. To check the expression condition, four 50 ml
cultures were prepared in the TB-medium containing 30 µg/ml kanamycin. The cultures grew under shaking at 37°C until OD = 0.8–1.0. The expression was started with
IPTG inducing at a concentration of 0.5 mM and performed either at 37°C or at 22°C
for 3 h as well as over night (ON).
The expression was checked with SDS-PAGE (Figure 2.1). The best condition for
the Syt1 expression was at 22°C for at least 3 h after IPTG inducing. After expression,
the culture was centrifuged (4000 RPM, 15 min, 4°C) and the pellet was resuspended
with a little amount of extraction buffer (all buffers used in this subsection see page 26,
Table 2.3). The resuspended cell was kept at -20°C.
kDa
170
130↑, 100↓
70
55
40
35
25
15
oo
Syt
oo
10
37°C 37°C
ON
3h
no
IPTG
22°C 22°C
3h ON
Figure 2.1: Expression test of Syt1 under different conditions. Each 50 ml TB-KANA-culture
(30 µg/ml) was incubated at 37°C until OD = 0.8–1.0. The expression was started
with 0.5 mM IPTG and performed under different conditions (shown on the gel
picture). The sample in the middle (“no IPTG”) was taken before adding IPTG.
Syt1 was purified with immobilized metal ion affinity chromatography (IMAC)
followed by Ion Exchange Chromatography (IEXC). The resuspended cell pellet was
thawed and filled with the same amount of extraction buffer with 10% sodium cholate.
lysozyme (f.c. 1 mg/ml), PMSF 1 mM), MgCl2 (f.c. 1 mM) and a few crumbs of DNAse
I were added to the cell solution and the mixture was incubated under stirring for
30 min at RT. The lysate was furthermore homogenized with 4 x 40 strokes using ultra31
sound (large Tip, 50% duty cycle, microtip limit) and incubated for 15 min at RT. The
solution was centrifuged at 15000 G for 30 min (SLA-1500).
The supernatant was carefully collected and mixed with 9 ml Ni-NTA-Agarose
slurry (for a culture of 6 l). The lysate-Ni-NTA mixture was incubated for 3 h at 4°C on
the rolling incubator. The mixture was filled into 50 ml Falcon tubes and centrifuged
at 3000 U for 10 min (Beckman J6-MI). The supernatant was removed into a bottle (Qiagen) and the Ni beads were resuspended in a small amount of supernatant and filled
into a large column. The filled column was washed with 300 ml of the washing buffer.
Finally, Syt1 was eluted with the elution buffer in five fractions (5 x 10 ml). The eluates
were directly mixed with DTT (f.c. 1 mM). After a rough concentration determination (Bradford) the concentrated fractions were pooled together and mixed with 50 µl
thrombin (1 U/µl) per 10 ml of volume to cleave the histidine tag. The protein solution
was dialyzed in at least 10 times the volume of sample with the desalting buffer at 4°C
overnight. All the purification steps and the thrombin cleavage itself were checked
with SDS-PAGE (Figure 2.2a).
Thrombin and other proteins were furthermore removed with IEXC using a MonoS
5/5 HR column on the Aekta purification system. After loading the protein solution
onto the column, elution was programmed as first a prewash with 70% buffer A and
30% buffer B (page 27, Table 2.3) for two column volumes, followed by increasing
buffer B’s fraction from 30% to 100% for five column volumes. Finally the column
was washed using 100% buffer B for two column volumes. All of these eluates were
retrieved as sets of separate fractions of 2 ml each. The concentrated fractions were
checked using SDS-PAGE (Figure 2.2b). The purification yielded ca. 1 mg protein per
6 l of culture.
32
a
b
kDa
170
130↑,100↓
70
55
40
35
25
kDa
70
Syt-His
Syt
oo
Syt
55
40
35
25
oo
15
15
10
10
Pell Sup Fl
W F2 F2* F3 F3*
F2* F3* IEXC
Figure 2.2: Purification of Syt1 with immobilized metal ion affinity chromatography and ion
exchange chromatography. a. Purification steps using Ni-NTA beads: the pellet after cell lysis (pell), the supernatant of the lysate (Sup), the flow-through after binding on the Ni-NTA beads (Fl), washing the column with the wash buffer (W), the
most concentrated fractions checked with Bradford Kit (F2 and F3) and the proteins after thrombin cleavage and dialysis (F2* and F3*). b. Purification with IEXC:
protein input (F2* and F3*) and IEXC output (IEXC).
33
2.2.2
Determination of the protein concentration
2.2.2.1
UV absorption
The concentrations of the SNARE proteins, Syt1 and α-SNAP were determined
with UV absorption spectroscopy (280 nm) [28,88]. The molar extinction coefficient ε of
a protein with amino acids sequence number n could be calculated as Equation 2.1 [43]:
280
280
280
ε280
Protein = nTyr · ε Tyr + nTrp · ε Trp + nCys · ε Cys
In water:
(2.1)
−1
−1
ε280
Tyr = 1490 mM cm
−1
−1
ε280
Trp = 5500 mM cm
−1
−1
ε280
Cys = 125 mM cm
The concentration of a homogenous protein solution with extinction coefficient ε
could be calculated using Lambert-Beer law (Equation 2.2), in which c is the protein
concentration and d is the thickness of the light path.
Abs = ε · c · d
(2.2)
The protein concentration was measured using a Nanodrop Spectrometer (ND-1000,
Thermo Scientific) and the extinction coefficients of the proteins were taken from the
program “ProtParam tool” [32].
2.2.2.2
Bradford assay
By protein purification with Ni-NTA beads (page 30, subsection 2.2.1) the concentration of the different fractions of the eluates were checked tentatively with a Bradford
Kit (Biorad) [15]. The Bradford assay is a sensitive photometric method for quantitative
determination of the protein concentration under a denaturized condition in µg/ml
range. Coomassie brilliant blue G-250 is a triphenylmethane dye which can mainly
34
bind to the alkaline amino acids and then turns from its cationic – to its anionic state.
The absorption can be observed at 595 nm. To qualitatively test the protein concentration, 10 µl of each protein solution were mixed with 800 µl ddH2 O and 200 µl Bradford
reagent 5x (Biorad, USA).
2.2.2.3
Modified Lowry-Peterson protein determination
For determination of the protein concentration of the SVs, the SV samples were
treated according to a modified Lowry-Peterson method [90]. Bovine serum albumin (BSA) was used as the standard in different concentrations in the range of 0–
40 µg. A series of different dilutions of the SV samples (1:2, 1:5, 1:10 and 1:20) and
the standards were filled up with ddH2 O to an end volume of 1 ml. The solutions
were incubated with 100 µl DOC (Deoxycholic acid, 0.15%) for 10 min at RT. TCA
(Trichloroacetic acid, 72%) was added to the solutions to precipitate the proteins. The
mixtures were incubated for 10 min on ice and centrifuged for 10 min at 13 000 rpm at
4°C. The pellets were dissolved and incubated in 750 µl Lowry solution III (page 28,
Table 2.3, same for the Lowry solution IV) and 250 µl ddH2 O for 30 min at RT. Finally,
the samples were mixed with 75 µl of Lowry solution IV and incubated for 45 min at
RT. The absorbance was measured at 750 nm using a Novaspec II photometer (Parmacia Biotech, Germany).This part of the experiment was performed by Dr. Saheeb
Ahmed (neurobiology department).
2.2.3
Determination of Rhodamine Green concentration with UV absorption spectroscopy
To determine the mouse SVs concentration using FCS, a standard fluorescence dye
Rhodamine Green (RG) of known concentration (10 nM) was used to analyze the data
by directly comparing particle numbers. RG was solved in PBS (Sigma) and the concentration of RG was determined by UV-absorption spectroscopy (Cary 5E, Varian).
The molar extinction coefficient of RG is ε RG = 75000 M−1 cm−1 at 502 nm [72]. The
35
concentration was calculated using Equation 2.2.
2.2.4
SDS-PAGE
The purification steps of proteins were checked using SDS-PAGE (Sodium dodecyl
sulfate polyacrylamide gel electrophoresis) [68]. The recipes for the tricine gels are
listed in Table 2.5 [108]. 3.4 ml separating gel mixture (10%) was poured between the
glass plates (0.8 mm separation) and ca. 1 ml collecting gel mixture was added on top
of the separating gel until the chamber was full. A comb with either 10 or 15 wells
was inserted directly into the collecting gel solution and could be removed after about
5 min, when the polymerization was finished. The protein samples were mixed with
5x sample buffer (see page 27, Table 2.3, same with the following buffers) and heated to
90°C for 10 min. Of each sample, 10 µl were loaded into the gel pockets. The gel tank
was filled with the cathode buffer in the middle and the anode buffer outside the gel.
It took about 25 min to pull the sample through the collecting gel using 60 Volts and
the electrophoresis was promoted at 120 Volts until the blue color of the samples ran
out of the gel. For staining, the gel was dipped into the coomassie staining solution,
heated for 5 s in a microwave and rocked for 5 min. To destain the gel, the coomassie
staining solution was carefully removed and the gel was incubated in the coomassie
destaining solution for another 5 min. The gel could be furthermore destained in water
overnight.
Component
Collecting gel
Separating gel
Acrylamide 30%
200 µl
1.66 ml
Gel buffer
375 µl
1.68 ml
ddH2 O
925 µl
570 µl
-
1.06 ml
TEMED
2 µl
3 µl
APS 10%
10 µl
25 µl
Glycerol 50%
Table 2.5: Tricine (Schägger)-gels.
36
2.2.5
Reconstitution of liposomes
2.2.5.1
Reconstitution of small liposomes
For the Syt1 tethering experiments, the liposome was reconstituted using size exclusion chromatography as described in earlier papers [105, 138] with the following
modifications. All the lipids components were solved in the chloroform/methanol solution (page 25, Table 2.3, the following buffers in this subsection analogously) and
mixed with their composition for the liposome. The lipid solution was dried to a lipid
film with nitrogen gas or with a rotary evaporator if larger amounts were used. The
lipid film was resolved in HPCholate5 buffer to a final lipid concentration of 27 mM.
16.7 µl lipid mixture were added with protein at a protein:lipid ratio of 1:1000 or 1:750.
The lipid protein mixture was filled with HPCholate1.5 buffer to a final volume of
50 µl. The liposomes were formed by detergent removal using a Sephadex G50 econo
column (GE Healthcare, Biorad). The running buffer for the column was HP150. The
collected liposome volume was about 250 µl. The radius of the liposome was about
20 nm [138]. Figure 2.3 shows a schematic of the size exclusion chromatography. All
the lipid compositions used in this study are listed in Table 2.6.
Figure 2.3: Size
exclusion
chromatography
using
sephadex G50 econo column to reconstitute
small liposomes.
The lipid mixture was
separated by the sephades 50G column to
three phases: the removed detergent (top),
the remaining free lipids and proteins (center)
and the formed liposomes (bottom).
2.2.5.2
Reconstitution of large unilamellar vesicles
Large unilamellar vesicles (LUVs) were prepared for fusion and docking experiments with purified chromaffin granules (CGs) by reverse phase evaporation and
37
Liposome
PC
PE
TRPE
OGPE
PS
Cholesterol
PiP2
TR
70
19
1
0
0
10
0
TRPS
50
19
1
0
20
10
0
TRPS5
65
19
1
0
5
10
0
TEPS12
58
19
1
0
12
10
0
OG
70
18.5
0
1.5
0
10
0
OGPS
50
18.5
0
1.5
20
10
0
OGPiP
49
18.5
0
1.5
20
10
1
Table 2.6: Lipid composition of the liposomes in the Syt1 tethering experiments. Numbers
are given in mol%.
extrusion through polycarbonate membranes with a pore size of 100 nm [23, 55,
89]. The lipid composition of the liposome was PC:PE:PS:Cholesterol:PI:PIP2:NBDDOPE:Rhodamine-DOPE = 45:12:10:25:4:1:1.5:1.5. The size of the LUVs was verified by
light scattering. The ∆N complex or the 2:1 complex (page 24, Table 2.2) were inserted
into the preformed LUVs at a protein:lipid ratio of 1:500 using detergent n-octyl-β-Dglucoside followed by a overnight dialysis [23, 55]. This reconstitution was part of the
investigation project of Dr. Yongsoo Park (neurobiology department).
2.2.6
Purification of mouse and rat SVs
Purification of mouse and rat synaptic vesicles was described previously [53, 85]
and the isolation procedure was optimized by Dr. Saheeb Ahmed in his PhD thesis [1].
This purification protocol allows a one brain isolation of SVs within 24 h and had an
optimized yield and purity.
After differential centrifugation the SVs were separated from the brain and isolated by means of size-exclusion chromatography (Sephacryl S-1000 Superfine HR, GE
Healthcare, Germany). Elution was monitored at a wavelength of 280 nm. The purity
of the SV isolation was checked by a dotblot assay [64]. The average size of the SVs was
38
determined by cryo electron microscopy and the average diameter of the SVs found
to be around 40 nm. The purified SVs were eluted using a HB-100 buffer (page 28,
Table 2.3) and the most concentrated fraction was used directly for the FCS measurement (see page 11, subsection 1.5.1) without centrifugation to determine the vesicle
concentration. All the mouse and rat SV purifications were performed by Dr. Saheeb
Ahmed.
2.2.7
Purification of chromaffin granules (CGs)
Purification of CGs was described previously [89]. CGs were separated from the
bovine adrenal medullae by differential centrifugation and were isolated and purified with a continuous sucrose density gradient from 0.3 M to 2.0 M (SW41T1 rotor,
27 000 rpm, 1 h). The purified CGs were collected from fraction 16 (pellet) and the
purity was checked by western blotting [89]. The purified CGs were resuspended in
CG buffer (see page 25, Table 2.3). The average diameter of CGs was about 167 nm
determined by the cryo-electron microscopy data. All the CGs were provided by Dr.
Yongsoo Park.
2.2.8
Two-photon confocal fluorescence microscopy setup
To simultaneously excite two different colors of liposomes, 22 mW of the output
power of a Chameleon titanium-sapphire laser (Coherent, USA) was used for twophoton excitation. The 800 nm, 87 MHz laser beam was expanded using a lens system
and coupled to an IX71 inverted microscope (Olympus, Germany), reflected to the
top of the microscope by a dichroic mirror DC1 (715 DSCPXR, AHF, Germany) and
focused by a UPlanSApo 60x/1.2w water immersion objective (Olympus, Germany).
The emitted photons passed through the objective and the dichroic mirror DC1.
Scattered light from the excitation beam was blocked by a two-photon rejection
filter 2P-SP (E700SP2, AHF, Germany). The emission was collimated using a second
lens system and reflected into the detection side of the setup using an ultrabroad band
39
dielectric mirror M (650–1130 nm, Newport, USA). The emission was furthermore split
by a second dichroic mirror DC2 (590 DCXR, AHF, Germany), filtered in each direction by a bandpass filter BP1/BP2 (HQ 645/75 and HQ 535/50; AHF, Germany) and
focused with a lens (either L1 or L2, respectively) to an avalanche photodiode (APD1
or APD2, respectively; SPCM-AQR-13, Perkin-Elmer, Canada; see Figure 2.4 a). The
TTL (transistor-transistor-logic) signals from the APD were analyzed using a 4-channel
router (PRT 400, PicoQuant GmbH, Germany) and a TCSPC (time-correlated single
photon counting) card (TimeHarp200, PicoQuant GmbH, Germany) and saved in PicoQuant’s TTTR (time-tagged time-resolved) format. The correlation was processed
using a home-made program (provided courtesy Matthias Grunwald).
Figure 2.4 b shows a schematic of the focus volume for two-photon excitation.
The black line indicates the laser beam, and the dark red ellipse marks the excited
effective volume Ve f f of two-photon excitation (TPE). Only the samples inside Ve f f
can be excited while the surrounding volume remains dark. The effective volume of
two-photon excitation in this study was approx. 0.3 fl.
40
b
a
Objective
APD 1
L1
BP 1
DC 1
BP 2
L2
Ti:Sa 800 nm
2P-SP
M
APD 2
DC 2
Figure 2.4: Confocal fluorescence microscopy setup for two-photon excitation. a. The pulsed
800 nm Ti:Sa laser beam is reflected to the objective using a dichroic mirror (DC1)
and the emission passes through the objective using the same pathway. The scattered light of the laser beam is blocked by a two-photon rejection filter (2P-SP). The
emission light is reflected to the detection side by an ultrabroad band dielectric mirror (M) and split by a second dichroic mirror (DC2) to finally arrive at one of two
avalanche photodiodes (APD1 or APD2) after having been filtered by a bandpass
filter (BP1 or BP2) and focused by a lens (L1 or L2). b. The dark red ellipse marks
the Ve f f of the TPE.
41
42
Chapter 3
Results
3.1
Cis- and trans-membrane interaction of synaptotagmin-1
Synaptotagmin-1 (Syt1) possesses two conserved calcium-binding domains–C2A
and C2B–, which serve as partial calcium coordination sites [37, 95]. The C2 domains
mediate calcium-dependent binding to both SNARE proteins [7, 19, 21, 125] and acidic
membrane lipids, with a specific affinity for the membrane lipid phosphatidylinositol4,5-bisphosphate (PiP2) [3, 6, 13, 42, 56, 59, 73, 95, 104, 134]. Separated from the calcium
binding pocket, the Syt1 also possesses a poly-basic lysine stretch in C2B domain,
which binds to anionic membranes containing phosphatidyl-serine (PS) or PiP2 in a
calcium-independent manner [6,19,73,104,134,141]. The 3D-structure of Syt1 is shown
in Figure 1.5.
However, it is not yet clear how exactly these binding activities contribute to the
acceleration of exocytosis [96]. Recent studies suggested that Syt1 may not only bind
trans to the plasma membrane but also cis to its own membrane [67, 111]. Using fulllength Syt1 and its mutants, in which either the Ca2+ binding C2 domains or the Ca2+
independent binding site in C2B domain were disrupted, a structural requirement of
the cis- and the trans-binding under the influence of PS/PiP2 and SNAREs was dis43
CHAPTER 3. RESULTS
sected with a sensitive tethering assay based on FCS and FCCS (see page 44, subsection 3.1.1). No SNARE proteins are inserted in the host membrane of Syt1, so that
tethering will be observed in a stable state without undergoing fusion.
3.1.1
Tethering assay based on FCS and FCCS
The tethering experiments were performed with a two-photon confocal microscope (page 39, subsection 2.2.8) and the data were analyzed using FCS (page 11,
subsection 1.5.1) and FCCS (page 14, subsection 1.5.2) techniques. Figure 3.1 shows
a schematic of the tethering experiment data. The different percentages (0%, 50%,
and 100%) in Figure 3.1 reflect the particle number of tethered liposomes Nx (page 15,
equation 1.11) divided by the total number of the green liposomes Ng (page 15, equation 1.10):
Tethering =
Nx
· 100
Ng
(3.1)
The result was corrected by substituting a small percentage of the tethering experiment
with Syt1-TR and OG liposomes containing no PS or, in the case of the binding assay,
the TRPS and OGPS liposomes in HP150 buffer, respectively.
G (t)
0.5
0.25
0
0% Tethering
50% Tethering
100% Tethering
t
Figure 3.1: Schematic of typical FCCS curves in the tethering assay. Red and green liposomes
are in 1:1 ratio. The red and green lines are the FCS measurement for each liposome
color and the blue line is the simultaneous FCCS measurement. All three liposome
pictures contain two red and two green particles.
44
CHAPTER 3. RESULTS
For the experiments, Texas Red (TR) and Oregon Green (OG) labeled small liposomes (page 37, subsection 2.2.5.1) were first measured with FCS at a 1:20 dilution
to check their concentrations. Using HP150 buffer (for all buffers see page 25, Table 2.3) the liposomes were diluted to a stock solution with about 100 liposomes in the
excitation volume for each color. The tethering experiment was performed in either
EGTA buffer or Calcium buffer. The experiment was started directly after mixing (vortex) 90 µl buffer and 5 µl of each the red and green liposomes at RT. Each droplet (20 µl)
measurement took 72 s and the data trace was split into six fragments for the correlation.Each measurement set was repeated at least once with fresh liposomes and buffer.
All the results were obtained by averaging of two fresh preparations with the error bar
marking the range of the respective data points obtained. Figure 3.2 shows two sets
of experimental data, one from an experiment under no-tethering conditions and on
from an experiment under full-tethering conditions.
a
b
0 ,4
0 ,4
( τ)
0 ,3
0 ,2
0 ,2
G
G
( τ)
0 ,3
0 ,1
0 ,1
0 ,0
0 ,0
1 E -6
1 E -5
1 E -4
1 E -3
τ( s )
0 ,0 1
0 ,1
1 E -6
1 E -5
1 E -4
1 E -3
τ(s )
0 ,0 1
0 ,1
Figure 3.2: Example FCCS data from the tethering experiments. a. Control experiment with
OGPS and TR liposomes without Syt in EGTA buffer (no-tethering conditions).
b. Tethering experiment with OGPS and Syt1-TR liposomes in calcium buffer (fulltethering conditions).
45
CHAPTER 3. RESULTS
3.1.2 Cis- and trans-interaction of Syt1 to the acidic lipid membrane
To investigate the membrane interaction of Syt1 (WT), full-length Syt1 and its mutants (a*B, Ab*, a*b* and KAKA) were reconstituted into red TR liposomes in a protein:lipid ratio of 1:1000. Tethering was measured with the protein-free green OG liposomes using FCCS under two-photon excitation. In the a*B and Ab* mutants, either
one or both C2 domains (a*, b*) were disrupted, so that one or both C2 domains cannot
bind calcium and thus to the acidic phospholipids. In the KAKA mutant, the polybasic
stretch of the C2B domain was mutated by two of the lysine residues, and the mutant
cannot bind to PiP2 or PS without presence of calcium. Figure 3.3 shows the domain
structures of Syt1 and the mutants used.
TM
1
58-79
C2A
140
C2B
265 271
421
Syt1* (1-421)
a*B (D178A, D230A, D232A)
Ab* (D309A, D363A, D365A)
a*b* (D178A, D230A, D232A,
D309A, D363A, D365A)
KAKA (K326A, K327A)
Figure 3.3: Domain structures of Syt1 and its mutants used in this experiment.
The TR liposomes contain 1 mol% TRPE and either 0 mol% (TR) or 20 mol%
PS (TRPS) whereas all the OG liposomes contain 1.5 mol% OGPE and 20 mol%
PS (OGPS). One population of the OG liposomes was reconstituted with an additional
1 mol% PiP2 (OGPiP). All of the liposome compositions are listed in Table 2.6, on
page 38.
The first set of experiments was tethering between the Syt1 (WT) bearing TR liposomes and protein-free OGPS liposomes in either the absence or the presence of
100 µM CaCl2 [111]. A tethering efficiency of 73% was observed with calcium whereas
the tethering was reduced to 20% without calcium (Figure 3.4a, WT). The experiments
were repeated with mutants of Syt1. a*B and Ab* showed similar results as the WT,
46
CHAPTER 3. RESULTS
i.e. calcium almost doubled the tethering to more than 85%. The double mutant
a*b* did not show a calcium dependence, whereas the KAKA mutant tethered the
OGPS liposome exclusively in the presence of calcium, as expected [3, 60, 141] (Figure 3.4a, a*B, Ab*, a*b* and KAKA). In agreement with previous observations, the
calcium-independent tethering is mediated by the polybasic lysine stretch of the C2B
domain [3, 141]. In all experiments with mutants, a small increase in tethering was
observed compared to the respective WT experiment under the same conditions. One
possible explanation for this increase is an enhanced membrane interaction due to the
removal of charges.
The KAKA mutant possesses only calcium-dependent binding sites and the binding was inhibited when calcium was chelated with 500 µM EGTA. To check the binding
specifity of the calcium, the tethering experiment with the KAKA mutant was repeated
with an extra 1 mM ATP/MgCl2 . Mg2+ did not influence membrane tethering. A control experiment was performed with protein-free TR liposome. No tethering was observed in both the absence and presence of calcium, which indicated that the tethering
was dependent on Syt1 only (Figure 3.4a, Control). Fusion was monitored by lifetime
analysis of OG. Because TR and OG form a FRET pair (page 17, subsection 1.5.4 [23]), a
reduction of the OG lifetime indicates lipid mixing of the liposomes. No major change
in lifetime was observed and thus it was concluded that fusion did not occur.
To analyze whether PiP2 enhances membrane tethering, the trans-tethering experiments described above were repeated using OGPiP liposome as target liposome (Figure 3.4b). No major differences were observed in all these tethering experiments. According to a previous study, a high percentage of PS could reduce the PiP2 effect in the
binding assay between the labeled Syt1 and the labeled liposomes [95].
To investigate whether cis-binding of Syt1 to its own membrane affects its tethering
activity, the entire tethering experiments in Figure 3.4 a and b were repeated using Syt1
bearing the TRPS liposome instead of the TR liposome. Most strikingly, the presence of
20% PS almost completely inhibited the membrane tethering under all conditions (Figure 3.4c and d).
47
CHAPTER 3. RESULTS
To examine the the lower boundary of this cis-binding effect, the PS percentage
in the host liposomes was lowered to 12% (TRPS12) and 5% (TRPS5). 12% PS is the
typical PS composition in a native rat synaptic vesicle [119]. Similarly, cis-preventing
was observed in the tethering experiment with Syt1 WT (Figure 3.5a). The cis-binding
could finally be inhibited using 5% PS in TR liposome in the presence of calcium or
PiP2 (Figure 3.5b). Here, the calcium-dependent trans-binding was recovered to the
maximum binding rate, whereas the calcium independent tethering only appeared if
the target membrane contained 1% PiP2.
These unexpected results suggested a different binding model between Syt1 and
liposomes compared with the findings of previous studies, which had shown that soluble C2AB domains were able to cluster liposomes containing acidic phospholipids in
the presence of calcium [3,60,141]. When comparing these results, it is conceivable that
membrane anchorage reduces the mobility of the C2 domains in such a way that cisbinding alone already exhausts both binding sites of C2A and C2B, and that therefore
no C2 domain is available for trans-binding. To shed light on this issue, we carried out
further tethering experiments using a soluble C2AB domain (97–421). These results are
laid out in the next subsection.
Figure 3.4: Liposome tethering mediated by membrane-anchored Syt1. Tethering was measured using TP-FCCS in the absence (black) or the presence (red) of 100 µM CaCl2 .
The donor liposomes contain either 0% or 20% PS, while the acceptor liposomes
contain either 20% PS or both 20% PS and 1% PiP2. a. Tethering experiment using
wild-type Syt1 (WT) and its mutants. Except for the a*b* mutant, calcium drastically increases the tethering rate in all the other cases. The KAKA mutants bind
acceptor vesicles only in the presence of calcium. b. Tethering experiment in the
presence of PiP2 in the acceptor vesicle confirming the results in the first set of the
experiments (a). c. Cis-binding mediated by PS in the donor vesicle. Tethering was
observed neither for Syt1 nor for any of its mutants. d. Cis-binding in the presence
of PiP2 in the acceptor vesicle could still be observed. All control experiments were
performed with protein-free donor vesicles.
48
CHAPTER 3. RESULTS
Tethering (%)
a
0
20
40
60
80
1 00
Donor vesicle
WT
Acceptor vesicle
Control
a*B
PS
Ab*
a*b*
KAKA
b
0
20
40
60
80
1 00
WT
Control
a*B
PS
PiP2
Ab*
a*b*
KAKA
c
0
20
40
60
80
1 00
WT
Control
a*B
PS
PS
PS
PS
PiP2
Ab*
a*b*
KAKA
d
0
20
40
60
80
1 00
WT
Control
a*B
Syt-1
Ab*
a*b*
- Ca2+
KAKA
+ Ca2+
Control: No Synaptotagmin in donor vesicle
49
CHAPTER 3. RESULTS
a
T e t h e r in g %
1 0 0
b
T R P S 1 2 %
7 5
n o C a
1 0 0 µ M
5 0
1 0 0
T R P S 5 %
7 5
C a
5 0
2 5
2 5
0
0
O G P S
O G P IP 2
O G P S
O G P IP 2
Figure 3.5: Syt1 mediated membrane tethering in the presence of 12% and of 5% PS in
the cis-membrane. Tethering was measured in the absence (black) and the presence (red) of 100 µM CaCl2 . a. 12% PS prevented tethering under all conditions.
b. With 5% PS in the Syt1 liposome membrane, tethering required either the presence of calcium or PiP2 in the target membrane.
3.1.3
The soluble C2AB domain of Syt1 is able to cluster liposomes
at a saturating concentration
The C2AB (97-421) fragment and its single cysteine mutant AF-C2AB labeled with
alexa fluor 488 were used to cross-link the TR and OG liposomes containing 20% PS.
The domain structures are shown in Figure 3.6.
TM
1
Syt1
(1-421)
C2AB
(97-421)
C2A
58-79
140
C2B
265 271
421
AF-C2AB (97-421, S342C
Alexa Fluor 488)
Figure 3.6: Domain structures of the soluble C2AB domain and its Alexa Fluor 488 labeled
mutant.
Different C2AB dilutions ranging from 50 nM to 860 nM were used to cluster the
green and red liposomes under 100 µM calcium. All experiments were started by
adding the C2AB domain into the liposome mixture. Clustering could be observed
only upon addition of elevated C2AB concentrations. The first evident clustering was
determined with 215 nM C2AB (Figure 3.7). No clustering was observed below this
50
CHAPTER 3. RESULTS
concentration even after 30 min of incubation. The clustering was reversible by adding
2 mM EGTA after clustering. The experiments with either EGTA or calcium alone, as
well as with C2AB in EGTA buffer are shown as negative controls (Figure 3.7).
40
T erh erin g %
30
PS
PS
PS
PS
EGTA
PS
20
10
PS
C2AB
Ca2+
0
EGTA / 1 mM
CaCl2 / 100 mM
C2AB / nM
+
o
o
o
+
o
+
o
430
o
+
50
o
o
+
+
107.5 215
o
o
+
430
+
860
2 mM
+
430
Figure 3.7: Clustering of liposomes mediated by soluble C2AB domain. Columns 1-3: negative controls. Columns 4-8: clustering experiment using soluble C2AB domain of
Syt1 of the concentration from 50 nM to 860 nM. Column 9: clustering mediated by
430 nM C2AB could be reverted using 2 mM EGTA.
It is essential to determine the concentration of soluble C2AB relative to the liposome required to cross-link the liposomes and to find the reason for the absence
of clustering at low concentrations. A binding experiment was performed in which
Alexa Fluor 488 labeled C2AB fragments (AF-C2AB) were added to the TRPS liposome containing 20% PS (Figure 3.8) and the binding was measured with FCCS, which
is capable of monitoring free and bound liposomes separately under the same condition. The lowest non-clustering concentration (50 nM) and the lowest clustering concentration (215 nM) of AF-C2AB were added to the TRPS liposome (Figure 3.8a). Both
experiments resulted in a high efficiency of soluble AF-C2AB binding to the liposome.
51
CHAPTER 3. RESULTS
All the results were calculated as the ratio of bound liposome relative to the entire liposome. At 215 nM, when clustering began to be observable, all the liposomes were
bound with AF-C2AB. The binding could be reversed by adding 1 mM EGTA.
a
PS
b
Bou nd of AF - C2AB %
Bo un d o f lip oso me s %
PS
PS
PS
100
50
0
EGTA / 1 mM
CaCl2 / 100 mM
AF-C2AB / nM
AF-C2AB
Ca2+
100
50
0
o
+
50
+
o
50
o
+
o
+
215 215
EGTA / 1 mM
CaCl2 / 100 mM
AF-C2AB / nM
C2AB / nM
o
+
50
o
+
o
50
o
o
o
o
+
o
+
+
+
215 215 50 50
o
o 400 150
Figure 3.8: Binding assay of soluble C2AB labeled with alexa fluor 488 to TRPS liposome
containing 20% PS. Binding was measured by TP-FCCS. a. Calculated as fraction
of the bound liposomes with AF-C2AB. b. Columns 1-4, calculated as fraction of
bound AF-C2AB fragments with liposome. b. Columns 5-6, binding was competed
by unlabeled C2AB fragments in two different concentrations.
The difference between calcium-dependent clustering and binding indicates that a
saturation of binding could be necessary to achieve clustering. To verify this finding,
the data from the very same experiments were used to calculate the fraction of bound
AF-C2AB fragments among the total number of AF-C2AB fragments (Figure 3.8b).
Contrary to the liposome binding rate, more than 60% of the fragments were bound
to the liposomes at 50 nM AF-C2AB whereas only 30% binding rate was achieved at
215 nM. The binding was calcium-dependent and could be inhibited with 1 mM EGTA
(Figure 3.8b, columns 1-4).
From this set of FCCS data, the fraction of bound liposomes and AF-C2AB, as
52
CHAPTER 3. RESULTS
well as the fraction of free components can be read out easily. With these, the binding
constant K D between liposomes ([ A]) an AF-C2AB ([ B])can be calculated. The binding
kinetics can be described by:
A + B AB
(3.2)
and the binding constant K D is then calculated as:
KD =
[ A] f ree · [ B] f ree
[ AB]
(3.3)
The total number of liposomes [ A]0 is a sum of the bound liposomes [ AB] and the
free liposomes [ A] f ree . The fraction of bound liposomes can be calculated as:
[ B] f ree
[ AB]
=
[ A ]0
K D + [ B] f ree
In this equation,
[ AB]
[ A ]0
(3.4)
can be taken directly from the fraction of bound liposomes in
Figure 3.8a (“Bound % of liposome”), and [ B] f ree can be calculated with the AF-C2AB
concentration and the fraction bound AF-C2AB in the Figure 3.8b (“Bound % of AFC2AB”). Using these data, the binding constant K D can be calculated as in Table 3.1:
[ B ]0
[ B] f ree
KD
(nM)
(nM)
0.6
20
20
0.3
150.5
16.7
[ AB]
[ A ]0
[ AB]
[ B ]0
50
0.5
215
0.9
(nM)
Table 3.1: Binding constant between liposomes and AF-C2AB fragments.
The average K D is about 18 nM. With Equation 3.4, the binding kinetics between
liposomes and AF-C2AB can be computed (Figure 3.9).
Addition of 215 nM AF-C2AB, which is 10-fold over the K D value, lead to a concentration of 150.5 nM free AF-C2AB fragments after binding to the liposomes. According to the binding kinetics, this high concentration is already within the saturation
range of the AF-C2AB. This was also confirmed by the binding experiments using extra amounts of the unlabeled C2AB fragments, in addition to the 50 nM AF-C2AB.
53
CHAPTER 3. RESULTS
1 ,0
[A B ]/[A ]0
0 ,8
0 ,6
K
D
= 1 8 n M
K
D
= 2 0 n M
K
D
= 1 6 .7 n M
0 ,4
0 ,2
0 ,0
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
3 0 0
[ B ] fre e ( n M )
Figure 3.9: Binding kinetics between liposomes (A) and AF-C2AB (B). [ AB] is the concentration of bound liposomes with AF-C2AB fragments. [ A]0 is the total concentration
of free and bound liposomes. The fraction of bound liposomes
[ AB]
[ A ]0
can be read off
from Figure 3.8a. [ B] f ree is the concentration of free AF-C2AB fragments, which can
be calculated as in the Table 3.1. K D is the binding constant between liposomes and
AF-C2AB fragments.
400 nM C2AB substantially competed with AF-C2AB and resulted in a binding rate
below 10%, whereas additional 150 nM C2AB reduced the binding rate to 35%, similar
as the binding rate using 215 nM AF-C2AB alone (Figure 3.8b, columns 5-6). In conclusion, soluble C2AB fragment clusters liposomes in a calcium-dependent manner and
requires the saturation of the binding sites on the membrane surface.
3.1.4
Syt1-SNARE interaction mediated membrane tethering
To investigate the membrane tethering mediated by Syt1-SNARE interaction [7,19,
21, 125] Syt1 and SNARE proteins were reconstituted separately into TR and OG liposomes. One TR liposome population contained 20% PS to check whether back binding
is preferred compared to the Syt1-SNARE interaction. All the SNAREs bearing OG liposomes contained no acidic lipids, so that the trans-tethering of Syt1 to PS or PiP2
was ruled out. Three different SNARE combinations were used in this set of experi54
CHAPTER 3. RESULTS
ments: Sx1A (syntaxin-1A, 183–288) alone, a binary 2:1 complex Sx1A-SN25 (syntaxin
1A, 183-288 and SNAP-25, 1-206) and a fully assembled ternary complex Sx1A-SN25Sb2 (synaptobrevin 2, 1–96, SNAP-25, 1-206 and syntaxin 1A, 183-288). The ternary
complex was reconstituted by incubating 2:1 complex and Sb 1-96 in a ratio of 1:2 at 4°C
over night, and the mixture was furthermore used for liposome reconstitution. Thus
the liposomes of the ternary complex also contained single Sx1A. The protein:lipid
ratio for Syt1 was 1:750 and for SNAREs it was 1:1000. All the experiments were performed in either EGTA or 100 µM calcium buffers.
Efficient tethering of about 40% was observed when Syt1 bearing liposome lacked
PS. The tethering was calcium independent except when using Sx1A alone. This result
agrees with the findings of previous studies [7,19,21,125], showing that the interaction
between Syt1 and Sx1A is enhanced by calcium (Figure 3.10a). Two negative controls
were measured using protein-free OG liposomes under the same conditions and with
an additional 1 µM free SNAP25 in the buffer, respectively, to exclude the unspecific
tethering caused by SNAP25 (Figure 3.10a, control).
The background tethering measured using TR-liposomes lacking Syt1 was significantly lower than in the tethering experiment (Figure 3.10b), indicating that tethering is mediated by Syt1-SNAREs interaction. The tethering was Syt1-SNARE specific,
since the presence of 10 mgl−1 BSA could not prevent it. Similarly to subsection 3.1.2,
tethering was missing completely when 20% PS was present in the Syt1 liposome (Figure 3.10c).
Since the Syt1-SNARE mediated membrane tethering mainly proceeded in a calcium-independent manner, the question arises whether this tethering is caused by the
polybasic stretch of the C2B domain of Syt1. Therefore, all the tethering experiments
were repeated using KAKA in which only the calcium-dependent binding sites C2A
and C2B are active (Figure 3.10d and e). Both tetherings mediated by SNARE complexes were reduced to the same levels as the non-tethering Syt1 (Figure 3.10d). The
calcium enhancement of the tethering mediated by single Sx1A was preserved, but
the tethering rate was slightly reduced as compared to the Syt1 WT experiment (Fig55
CHAPTER 3. RESULTS
Tethering (%)
0
a
No SNAREs (control)
Sn25 (control)
Sx1A
Sx1A-SN25
Sx1A-SN25-Sb2
No SNAREs (control)
Sn25 (control)
Sx1A
Sx1A-SN25
Sx1A-SN25-Sb2
b
c
20
40
60
80
100
Donor vesicle
Acceptor vesicle
2+
- Ca
SNAREs
2+
+ Ca
Sx1A
Sx1A-SN25
Sx1A-SN25-Sb2
- Ca
Sx1A
Sx1A-SN25
Sx1A-SN25-Sb2
+ Ca
Sx1A
Sx1A-SN25
Sx1A-SN25-Sb2
- Ca
2+
2+
SNAREs
2+
PS
d
e
Sx1A
Sx1A-SN25
Sx1A-SN25-Sb2
+ Ca
Sx1A
Sx1A-SN25
Sx1A-SN25-Sb2
- Ca
Sx1A
Sx1A-SN25
Sx1A-SN25-Sb2
+ Ca
Sx1A
Sx1A-SN25
Sx1A-SN25-Sb2
- Ca
2+
SNAREs
2+
2+
SNAREs
2+
PS
Sx1A
Sx1A-SN25
Sx1A-SN25-Sb2
2+
+ Ca
56
SNAREs
CHAPTER 3. RESULTS
Figure 3.10: Tethering of liposomes mediated by membrane-anchored Syt1-SNAREs interactions. Donor vesicles containing either 0% or 20% PS were reconstituted with
Syt1 WT or its KAKA mutant (P:L = 1:750). Acceptor vesicles free of PS were reconstituted with purified recombinant Sx1A (gray), a binary 2:1 complex Sx1ASN25 (red), or a ternary SNARE complex Sx1A-SN25-Sb2 (blue, P:L = 1:1000).
Control measurements were performed either without SNARE or with 1 µM soluble SNAP-25. Tetherings were carried out in the absence (− Ca2+ ) or presence (+ Ca2+ ) of 100 µM Ca2+. a. Tethering was observed under no PS condition. Ca2+ dependency was observed only by acceptor vesicle with Sx1A. b. Background tethering without Syt1 as a control. c. No tethering was determined
when Syt1 liposomes contained 20% PS which indicated back binding of Syt1.
d and e. Repeated tethering and back binding measurements with KAKA mutant
and the determined tetherings reduced to the background level except with Sx1A
liposomes which still showed a Ca2+ dependency.
ure 3.10d). Presence of PS in the Syt1 liposome significantly reduced the tethering with
single Sx1A to the level of the Syt1-lacking background (Figure 3.10e). This KAKA
mutant experiment was performed by Sabrina Schroeder and Chao-Chen Lin.
All the results reported in this section were summarized in a recent publication [123].
3.2
Characterization of mouse synaptic vesicles by average protein mass
Neuronal exocytosis requires fusion of synaptic vesicles (SVs) with the presynaptic
membrane. Characterization of the SVs in terms of protein contents and lipid composition as well as SNARE-mediated membrane fusion and neurotransmitter uptake is
essential to understand the fusion process and the storage of neurotransmitters. Biological and physical characterization of the rat SVs was done by Takamori et al. in
57
CHAPTER 3. RESULTS
2006 [119]. A detailed picture of the SVs including the quantified mass, composition
and copy number of the respective proteins and lipids as well as the size and density
per single vesicle was obtained (Figure 3.11).
Figure 3.11: Molecular model of an average SV. [119]
The mouse is one of the standard species of animal experiments and shares common ancestry with the rat. Thus, it is a natural extension of the previous study, to
characterize the protein and lipid structures of the mouse SVs at a single vesicle level.
In the following subsections, a vesicle counting assay based on TP-FCS is described
to determine the SV particle concentration, which is essential for characterizing all the
biological properties of a single vesicle. Furthermore, combining the protein concenFigure 4.determined
Molecular Model of
an means
Average SVof a modified Lowry-Peterson method, an average mass
trations
by
The model is based on space-filling models of all macromolecules at near atomic resolution.
(A) Outside view of a vesicle.
of proteins
persectioned
mouse
compared
that components
of the rat
could
be calculated.
(B) View of a vesicle
in theSV
middle
(the dark-coloredto
membrane
represent
cholesterol).
(C) Model containing only synaptobrevin to show the surface density of the most abundant vesicle component.
In addition to the proteins listed in Table 2, the following proteins were included (copy number in parentheses): VAMP4 (2), SNAP-29 (1), vti1a (2) syntaxin 6 (2), other syntaxins (4), other synaptotagmins (5), other Rab proteins (15), Munc-18 (2), other transporters (2), chloride channels (2), and trimeric
GTPases (2) (see Table S2 for more details).
58
Cell 127, 831–846, November 17, 2006 ª2006 Elsevier Inc. 841
CHAPTER 3. RESULTS
3.2.1
Synaptic vesicle labeling with FM 1-43 dye
FM dyes are widely used as outer membrane labeling dyes in monitoring endoand exocytosis, vesicle trafficking, and vesicle fusion in various systems [2, 12, 16, 54,
124]. FM 1-43 and its derivatives harbor a water-soluble cationic head group and a
membrane permeable hydrophobic tail, while their double bond central region carries
the fluorescent properties. The membrane association of the FM dyes is reversible.
FM 1-43 (N-(3-triethylammoniumpropyl)-4-(4-(dibutylamino) styryl) pyridinium
dibromide) is almost non-fluorescent in an aqueous solution but shows an increased
fluorescence of more than 40-fold intensity upon binding to liposomes [97].
To characterize the mouse SVs, they were tagged in the FCS measurement using
the FM 1-43 dye. The labeling protocol was modified from the previous paper [119].
10 µM FM 1-43 dye was mixed with the same volume of purified SV solution, which
was collected directly from size-exclusion chromatography, so that the final concentration of the FM dye was 5 µM. Incubation was performed for at least 2 min on ice.
Figure 3.12 shows two fluorescence traces taken from two measurements of 10 s each,
with either 5 µM FM 1-43 alone or FM-SV-mixture, in which FM dye had a final concentration of 5 µM and SV sample was 1:1 diluted by mixing with the FM dye solution.
The SVs were clearly labeled with the FM dye.
3.2.2
Concentration determination of SVs using TP-FCS
The concentration of the SVs was determined using two photon confocal microscopy at 40 mW and calculated with fluorescence correlation technique (FCS). The
experimental setup was as described previously (page 39, subsection 2.2.8) with a few
modifications. Only one APD detector (APD 2) was used to collect either RG or FM 143 fluorescence. Hence, a second dichroic mirror (DC 2) was not included for this
measurement, but instead Bandpass filters HQ535/50 as well as HQ645/75 were used
to select the emission of either RG or FM dye.
59
CHAPTER 3. RESULTS
F M
F M
2 0 0
1 -4 3 (5 µ M )
1 -4 3 (5 µ M ) + S V (1 :1 )
C o u n ts
1 5 0
1 0 0
5 0
0
0
2
4
6
8
1 0
T im e ( s )
Figure 3.12: SV labeling with FM 1-43 dye. SVs were incubated with the FM 1-43 dye for
2 min on ice with a final concentration of 5 µM FM dye (red); a background trace
of 5 µM FM 1-43 is shown as comparison (black).
The most concentrated SV fraction collected from the size exclusion chromatography (see page 38, subsection 2.2.6) was freshly labeled with FM 1-43 membrane
dye (see page 59, subsection 3.2.1). The homogeneity of the SVs was controlled by
measuring diffusion time. At least two SV dilutions with a final FM concentration of
5 µM were measured (Figure 3.13), and each dilution was measured at least twice with
fresh droplets. Every measurement took 6 × 10 s. The particle number of SVs in the
effective volume was calculated with equation 1.6, by NSV = 1/G (0).
A standard RG fluorescence dye solution with a concentration of 10 nM (page 35,
subsection 2.2.3) was used to calculate the concentration of the SVs. With the particle
NSV
number NRG of RG the concentration of SVs could be calculated as cSV = 10 nM· N
.
RG
The final concentration was an average of all the dilutions and the error bar was taken
to be the standard deviation of all the measurement fragments combined. The data set
will be shown in subsection 3.2.4 in combination with the protein concentration.
60
CHAPTER 3. RESULTS
0 ,8
D ilu
D ilu
D ilu
F M
1 :2
1 :4
1 :8
3 (5 µ M )
0 ,4
G
( τ)
0 ,6
tio n
tio n
tio n
1 -4
0 ,2
0 ,0
1 E -5
1 E -4
1 E -3
0 ,0 1
0 ,1
τ(s )
Figure 3.13: FCS measurement of rat SVs in three dilutions compared to FM 1-43 background. All three dilutions were measured in the presence of 5 µM FM 1-43 (red,
green and blue line). 5 µM FM 1-43 is shown as background (black).
3.2.3
Background correction due to the dark FM fluorescence
Although the free FM 1-43 dye is almost non-fluorescent [97], its emission at the
high concentration used in this experiment was still detectable. To eliminate this
“dark” background, a new calculation of the correlation function considering the experimentally observed fluorescence I exp (t) was developed.
The “dark” emission of 5 µM FM 1-43 was approximately a constant f independent
of time. Assuming that only a small part of the excess FM 1-43 dye was attached to
SVs, the experimentally observed fluorescence intensity I exp (t) could be calculated as
the sum of the labeled SV fluorescence I (t) and the FM’s near constant emission f . The
average of the experimentally obtained fluorescence could then be written in the form
61
CHAPTER 3. RESULTS
of a sum of the average fluorescence of labeled SVs h I (t)i and the FM constant f :
I exp (t) = I (t) + f
(3.5)
h I exp (t)i = h I (t)i + f
(3.6)
According to equations 1.1 and 1.2, the fluorescence correlation function can be written
as:
h( I (t) − h I (t)i) · ( I (t + τ ) − h I (t)i)i
(3.7)
h I (t)i2
Combining all three preceding equations, the experimentally determined correlation
G (τ ) =
is related to the pure labeled SV signal by the following equation:
h( I exp (t) − h I exp (t)i) · ( I exp (t + τ ) − h I exp (t)i)i
h I exp (t)i2
h(( I (t) + f ) − (h I (t)i + f )) · (( I (t + τ ) + f ) − (h I (t)i + f ))i
=
(h I (t)i + f )2
G exp (τ ) =
(3.8)
The numerator of the function results in the same as in equation 3.7, and G exp (τ ) can
be cast into a new form of G (τ ) multiplied with a constant factor:
G
exp
(h I exp (t)i − f )2
(τ ) = G (τ ) ·
= G (τ ) ·
h I exp (t)i2
(h I (t)i + f )2
h I (t)i2
(3.9)
Thus, the autocorrelation G (τ ) of labeled SVs can be calculated with the experimentally obtained correlation G exp (τ ) and a correction factor
G (τ ) = G exp (τ ) ·
h I exp (t)i2
(h I exp (t)i − f )2
(3.10)
In the experiment, the average fluorescence intensity h I exp (t)i is measured in units
of photon counts per second. 5 µM FM 1-43 was measured separately to obtain the
background constant f . The calculated correction factor to G exp (τ ) was determined to
be approximately 1.5.
3.2.4
Comparing the mass of proteins per vesicle of mouse and rat
SVs
If the particle concentrations of SVs were known, many biological properties could
be calculated at the level of a single vesicle. Combining the protein concentration of
62
CHAPTER 3. RESULTS
the same SV sample determined by using a modified Lowry-Peterson method (page 35,
subsection 2.2.2.3), the average protein mass per SV could be calculated. Figure 3.14
shows the plotted data of the vesicle concentration versus the protein concentration of
the same sample. Besides the measurements for mouse SVs (Figure 3.14, black), two
preparations of rat SVs purified by the same method were also measured (Figure 3.14,
red) for a direct comparison with the literature [119]. The mass of proteins per vesicle
was calculated as
m Pro =
c Pro
cSV · NL
in which c Pro is the protein concentration, cSV is the vesicle concentration, and NL is
Avogadro’s number.
1 2
M o u s e
R a t
V e s ic le c o n c . ( n M )
9
6
3
0
0
3 0
6 0
9 0
1 2 0
P r o te in c o n c . ( n g /µ l)
Figure 3.14: SV concentration plotted versus the protein concentration of the same sample.
Besides mouse SVs (black), two rat SV samples purified using the same method
(red) were plotted as a direct comparison.
The average vesicular protein mass of mouse SVs was calculated as the weighted
average of all the single purifications: m Pro = (21.4 ± 2.1) × 10−18 g/vesicle, corresponding to 12.9 ± 1.3 MDa. The average mass for rat SVs obtained from two purifications was (18.8 ± 3.2) × 10−18 g/vesicle, corresponding to 11.3 ± 1.9 MDa (Figure 3.15)
63
CHAPTER 3. RESULTS
which reflected the previous study measured with scanning transmission electron microscopy (STEM) [119].
tein mass for mouse was m Pro =
(21.4 ± 2.1) × 10−18
g/vesicle
(black), whereas for rat it was
(18.8 ± 3.2) × 10−18 g/vesicle (red).
3.3
1 0
5
0
R a t
one single SV, the average pro-
1 5
M o u s e
For
2 0
-1 8
for mouse compared to rat.
P r o te in ( 1 0
Figure 3.15: Calculated protein mass per SV
g /v e s ic le )
2 5
α-SNAP inhibits SNARE-mediated membrane fusion
but not partial SNARE-zippering
Neuronal SNARE proteins synaptobrevin 2 (Sb), syntaxin 1A (Sx) and SNAP 25A
(SNAP25) assemble from their N-termini toward their C-termini, pull the synaptic
vesicle membrane and the presynaptic membrane together and and provide free energy, which is essential to open the fusion pore [63, 79, 91, 116]. The disassembling of
the SNAREs requires AAA+ ATPase NSF (N-ethylmaleimide-sensitive factor) [81, 115]
and its co-factors SNAPs (soluble NSF attachment proteins), which attach NSF to the
SNARE complex, and proceeds via an enzymatic activity of NSF [22].
α-SNAP is one of the most abundant isoforms of SNAP proteins [94]. It possesses
an α1 region at its N-terminus which is essential for binding to the Q-SNARE complex,
so that deletion of the N-terminus completely disrupts the SNARE association [52].
Furthermore, α-SNAP consists of a hydrophobic loop at the N-terminus which is responsible for membrane attachment [132] and shows SNARE-independent membrane
lipid binding [110]. This membrane attachment also facilitates the disassembly of the
64
CHAPTER 3. RESULTS
SNARE complex [132].
Besides its regulatory function during membrane priming and fusion in cooperation with NSF [18, 83, 120, 133], free α-SNAP inhibits vesicle fusion and thus the exocytosis in various biological systems [4,9,103,120,127]. However, the details of this effect
are yet unclear.
In a previous study of Y. Park and coworkers (unpublished data), fusion between
purified chromaffin granules (CGs) and reconstituted large liposomes in the present
of α-SNAP was investigated. CGs are one of the different kinds of secretory vesicles
isolated from the chromaffin cells of bovine adrenal medullae and are also classified
as large dense-core vesicles (LDCVs), since they have an average size of about 200 nm
and consist of an electron-dense core [26]. For their fusion process, CGs contain the
same synaptobrevin and synaptotagmin as synaptic vesicles and are used as a model to
study SNARE-mediated exocytosis [86]. The large liposomes had an average diameter
of about 100 nm and were reconstituted with either 2:1 complexes (2 Sx:1 SNAP25) [35]
or ∆N complexes (Sx, SNAP25 and Sb49−96 ; see page 24, Table 2.2). The ∆N complex is
a stabilized Q-SNARE complex using a small Sb49−96 fragment which can be displaced
during assembly with the Sb of the target membrane and thus promotes rapid membrane fusion in vitro [91]. Besides α-SNAP, two N-terminal mutants were also used to
determine the functional requirements. The mutants were α-SNAP33−295 with deleted
N-terminus and α-SNAPF27,28S , which has two point mutations on hydrophobic loop
between the α1 and α2 helices at the N-terminus.
The fluorescence dequenching spectroscopy and the fluorescence anisotropy data
confirmed that 1 µM α-SNAP clearly inhibited CG fusion by blocking the SNARE assembly. Surprisingly, deletion of the Q-SNARE binding N-terminus (α-SNAP33−295 ) or
the membrane binding loop (α-SNAPF27,28S ) both revoked the inhibition, suggesting
that the fusion inhibition of α-SNAP requires Q-SNARE binding and membrane attachment. Furthermore, the inhibition function of the α-SNAP is dependent on membrane curvature, so that an inhibition of fusion was not observed with liposomes with
a diameter below 100 nm. In addition, the SNARE assembly during the CG fusion
65
CHAPTER 3. RESULTS
was checked using SDS-PAGE in the presence of α-SNAP and its mutants. Using the
neurotoxin TeNT, which cleaves only the free membrane anchored Sb, the detectable
Sb on the SDS gel indicated the SNARE assembly [51, 87]. The SNARE assembly was
not fully interrupted by α-SNAP and was recovered in the presence of α-SNAPF27,28S
which is in agreement with the results of the fusion experiment.
To answer the question whether the α-SNAP blocks the CG fusion but not the
SNARE assembly, a set of SNARE-dependent docking experiments using FCS was
performed under the same conditions as the CG fusion experiment described above.
A diffusion time change [5] of the liposomes upon binding with CGs indicated that
α-SNAP did not interrupt the CG docking and thus the SNARE zippering. In conclusion, α-SNAP inhibits the CG fusion via binding to SNARE complexes with the
N-terminus and attaching membranes using the hydrophobic loop, thereby clamping
the full SNARE assembly (Figure 3.16).
CG
Sb2
a1
SNAP25A
C
N
N
C
Sx1A
loop
a-SNAP
Liposome
Figure 3.16: Schmatic of CG docking with 100 nm liposomes in the presence of α-SNAP.
α-SNAP binds to the SNARE complex with its N-terminus and attaches to the
membrane via its hydrophobic loop. α-SNAP inhibits CG fusion but not the partial
SNARE assembly.
66
CHAPTER 3. RESULTS
3.3.1
Determination of CG docking by observing variations in diffusion time
The docking experiments were performed between ∆N complex anchored 100 nm
large liposomes (page 37, subsection 2.2.5.2) and purified CGs (page 39, subsection 2.2.7). Here, docking and fusion could not be distinguished, since only the liposome was labeled with both NBD and rhodamine dye, and only the red detection
channel was used. Therefore, docking mentioned in this subsection means all the cases
where liposome and CG are “gluing” together.
Docking is mediated by SNARE assembly. Using FCS measurements, the changes
in liposome diffusion time caused by a size increase upon docking could be determined. The FCS setup was described in a previous section (page 59, subsection 3.2.2)
using an HQ645/75 bandpass filter. The average liposome number in the effective volume was about 0.4, so that single particle docking events could be observed. For the
measurement, both 50 µg CGs and 10µl liposome were diluted in 1 ml CG buffer containing K-glutamate and MgCl2 (page 25, Table 2.3). α-SNAP was added with a final
concentration of 1 µM. The mixture was incubated for 5 min at 37°C. As a negative
control, 2 µM soluble Sb1−96 was pre-incubated with liposomes at 37°C for 30 min,
then mixed with CGs, and incubated for another 5 min. Sb1−96 blocks the full SNARE
assembly of the ∆N complex with the full-length Sb.
Figure 3.17 shows the FCS data of selected docking experiments. Three background diffusion time measurements of each ∆N liposomes, protein-free empty liposomes, and empty liposomes in the presence of CG yielded similar diffusion times of
about 10 ms in every case. The negative control experiment with soluble Sb1−96 in
the mixture of liposome and CG did not show any changes in diffusion time, as expected (Figure 3.17 first legend session, bottom left). An increased liposome diffusion
time of about 50 ms was observed in both docking experiments in the presence and
absence of α-SNAP (Figure 3.17 second legend session, top right).
Each docking experiment was measured in 30 × 10 s fragments, and the results
67
CHAPTER 3. RESULTS
L ip o s o m e s + C G s
L ip o s o m e s + C G s
+ α- S N A P
N o r m a liz e d G
( τ)
1 ,0
0 ,5
L ip
E m
E m
L ip
+ S
0 ,0
0 ,1
o s o
p ty
p ty
o s o
b 1 -9
m e s
lip o s o m e s
lip o s o m e s + C G s
m e s + C G s
6
1
1 0
1 0 0
1 0 0 0
τ( m s )
Figure 3.17: Diffusion time determination of the liposome docking experiments with CGs
using FCS. Liposomes were reconstituted with ∆N complexes. Empty liposomes
contained no proteins. The first data set with four FCS curves shows two liposome
measurements (black and red), and two negative control measurements, in which
either the empty liposomes were mixed with CGs (pink), or the liposomes were
mixed with CGs in the presence of soluble Sb1−96 (green). All these no-docking
measurements resulted in a diffusion time of about 10 ms. For the positive control
measurement, in which ∆N liposomes were mixed with CGs (blue), a diffusion
time of 50 ms was measured, indicating that liposomes were docked with CGs.
Addition of α-SNAP nearly did not change the diffusion time of liposomes (dark
blue), suggesting that α-SNAP cannot block CG-docking. Except for the two “liposome only” measurements, all the Liposome- CG mixtures were incubated at 37°C
for 20 min. Sb1−96 was pre-incubated with liposomes at 37°C for 30 min.
68
CHAPTER 3. RESULTS
are presented as a histogram of the diffusion time (Figure 3.18). In the cases of the
empty liposomes and the negative control experiments, almost all of the 30 measurement fragments showed a diffusion time within 15 ms, which indicated that there was
no docking. The docking experiment with ∆N liposomes and CGs showed a clearly
broader distribution of the diffusion time (Figure 3.18).
80
Empty liposome
Frequency (%)
ΔN-liposome + CG
ΔN-liposome + CG
+ Syb1-96
60
40
20
0
0
0.00
10
0.01
20
0.02
30
0.03
Diffusion
timetime
(ms)
Diffusion
40
0.04
50
0.05
Figure 3.18: Histogram of the docking experiment using liposome and CG. 30 × 10 s measurement fragments are plotted as a histogram of diffusion time. Protein-free
empty liposomes showed a narrow distribution of the diffusion time within 15 ms
(blue). Repeated fusion experiments with ∆N liposomes and CGs showed several
occurrences of diffusion times of up tp 50 ms indicating that a large fraction of
the liposomes was docked/fused (black). The negative control experiment performed through pre-incubating with soluble Sb1−96 of the ∆N liposomes showed
only little docking/fusion to CGs confirming that docking is mediated by SNARE
assembly (red).
69
CHAPTER 3. RESULTS
3.3.2
α-SNAP does not fully block CG docking
Following the FCS measurement, 15 ms were set as the criterion for discriminating
docked and free liposomes (Figure 3.18, dash line). The docked liposome (> 15 ms)
were calculated as percentage of the total of 30 measured events. Each measurement
was repeated at least five times using new preparations and the docking rate was obtained as the average and SEM (standard error of mean) of all the measurements (Figure 3.19).
About 65% of the liposomes were docked (fused) with CGs (Figure 3.19, no addition), and the deletion in the hydrophobic loop α-SNAPF27,28S did not affect the docking (fusion) which confirmed the previous fusion experiment with CGs (Figure 3.19,
F27,28S). The negative control experiment with soluble Sb1−96 showed that docking
is mediated by SNARE assembly (Figure 3.19, Sb1−96 ). Surprisingly, α-SNAP did not
block the docking (about 50%, Figure 3.19, α-SNAP), whereas in the fluorescence dequenching assay α-SNAP prevented the CG fusion completely. Docking was reduced
to background level using Sb1−96 , indicating that docking in the presence of α-SNAP
is SNARE mediated.
In conclusion, α-SNAP is able to inhibit CG fusion via its SNARE binding Nterminus and the membrane attaching loop, but α-SNAP does not inhibit partial
SNARE assembly and thus liposome docking.
70
CHAPTER 3. RESULTS
Docked liposomes (%)
80
70
60
50
40
30
20
No addition
Sb1-96
α-SNAP α-SNAP F27,28S
Sb1-96
Figure 3.19: α-SNAP does not block the SNARE-mediated CG docking. Docking was measured between ∆N liposomes and CGs with 5 min incubation at 37°C. The docking rate was calculated as percentage of all the 30 measurement fragments. The
error is given by SEM of more than five independent measurements. The normal docking/fusion rate was about 65% without adding α-SNAP (black). Docking
was found to be reversible using soluble Sb1−96 . This indicates that docking is
mediated by SNARE assembly (red). Docking was not fully disrupted by α-SNAP
(green) and was still reduced to the background level using soluble Sb1−96 (blue).
Deletion of the hydrophobic loop of α-SNAP had no effect on CG docking (light
blue), confirming the dequenching fusion experiment.
71
CHAPTER 3. RESULTS
72
Chapter 4
Discussion
The conflict of cis- and trans-membrane interaction
4.1
of synaptotagmin-1
Focusing on the two calcium dependent binding domains C2A and C2B of synaptotagmin-1, as well as its calcium independent binding site (polybasic lysine stretch) of
C2B, the contribution of each of these independent binding regions to the Syt1 mediated membrane tethering was investigated using a sensitive tethering assay based on
FCCS.
In this study, all the three binding sites engage in the tethering process of the membrane anchored Syt1 to the target membrane containing 20% PS or both 20% PS and
1% PiP2. Less than 50% tethering caused by the polybasic stretch occurs in the absence
of calcium whereas 100 µM CaCl2 can almost double the tethering rate with at least one
active C2 domain. This result confirms figures from the previous studies using isolated
C2 domains [3, 6, 95]. Membrane anchorage does not affect the trans interaction of the
C2 domains (Figure 4.1a). Similarly, Syt1 also tethers the membrane anchored SNARE
complexes. However, the binding is mainly based on the polybasic stretch of C2B and
does not show a significant dependency on calcium (Figure 4.1b). Particularly, binding
to the membrane anchored single Sx1A is slightly enhanced by calcium, in agreement
73
CHAPTER 4. DISCUSSION
with previous reports [7, 19, 21, 125].
a
Ca2+
PS
PS
b
Figure 4.1: Model of the trans-tethering of the membrane anchored Syt1 to the target membrane containing either acidic phospholipids or SNAREs. a. Membrane anchored
Syt1 tethers the target liposome in both calcium dependent and independent manners. Target membrane contains either 20% PS or both 20% PS and 1% PiP2. b. Syt1
binds the membrane anchored SNAREs without any calcium enhancement except
the binding with the isolated Sx1A.
Surprisingly, 20% PS in the host membrane of Syt1 completely inhibits the transtethering to the target membrane in all the cases, indicating that Syt1 interacts in cis
to its host membrane (Figure 4.2). In contrast to this finding, both soluble C2AB fragment and single C2B domain were reported to be able to cluster the vesicles in several
studies [3, 111, 141].
It is conceivable that clustering requires at least two independent binding sites.
The experimental finding in this study using soluble C2AB fragments confirms this
finding. However, clustering is only observable if C2AB fragments are saturated on
the liposome surface (Figure 4.3), which reflects the case investigated in previous studies. Below the concentration at which binding is saturated, C2AB binds only to one
liposome.
There is no direct explanation from the experimental results, why C2AB only clusters liposomes under saturating conditions, since all the liposomes in the solution are
similarly occupied. Nevertheless, the space and orientation requirements for placing
74
a
Ca2+
PS
PS
PS
b
PS
Ca2+
PS
PS
Figure 4.2: All trans-tethering is inhibited if the host liposome of Syt1 contains 20% PS. Target liposome contains either acidic phospholipids (a) or SNAREs (b). No tethering
is observable even in the presence of 100 µM CaCl2 .
C2AB
PS
PS
PS
PS
Ca2+
Figure 4.3: C2AB fragment is able to cluster the liposome under a saturating concentration
on the liposome surface. Clustering is reversible by adding 2 mM EGTA.
75
both C2 domains of one C2AB fragment on the membrane surface may still leave a
smaller area which is barely sufficient to bind one C2 domain, and furthermore to
cross-link a second liposome. Alternatively, cross-linking is only observable when the
liposome surface is fully covered by C2AB fragments, so that a maximum number of
C2AB bridges are achieved to stabilize the membrane tethering. With these findings,
many seemingly contradictory reports in the literature could be harmonized [3, 60].
In conclusion, the cis-binding model confirms the previous studies, in which membrane anchored Syt1 reduces the fusion efficiency in the in vitro experiments [71, 111],
and exhibits a declaration for this observation, that the C2 domains of Syt1 bind in cis
orientation to its host membrane containing acidic phospholipids and become inactive
for cross-linking the plasma membrane. Some of the conflicting reports about Syt1 in
artificial systems may be thus explained [67, 71, 111].
Along with this study, similar results have been reported [65] based on a related
experimental design [23], which are largely in agreement with this study. Complementing data has also shown that SNAREs and Syt1 involved in membrane fusion is
only calcium stimulated, if excess PS is present in the target membrane of Syt1 [70];
which is in accord with the tethering experiments using low percentage PS (5%) in the
host liposome of Syt1.
Most importantly, the results have emphasized the conflict of cis- vs.
trans-
membrane interaction of Syt1. Cis binding of the membrane anchored Syt1 is still observable even if the PS composition is lowered to 12% in the host membrane, which
is comparable to the natural synaptic vesicles containing more than 15% anionic phospholipids, suggesting that cis-interaction may occur under physiological conditions.
This has been confirmed by several preliminary data using SVs and CGs (unpublished
data and [58]). It is conceivable that in a docked intermediate, the vesicle membrane
and the plasma membrane get close enough to compensate for the cis-binding, and
one of the C2 domains can trans bind the plasma membrane – probably C2B due to its
special ability to PiP2 [6, 73, 104, 134, 141] – and thus mediate the cross-linking [67, 141].
Alternatively, cis-binding may be prevented by some other factors, such as molecular
76
crowding or charge screening. The latter has recently been observed [89], i.e. that 5 mM
polyvalent anions ATP can shield the cis-membrane interaction, since the calcium dependent membrane affinity of Syt1 is similar to the calcium affinity of ATP, and ATP
competes with the lipid membrane for binding Syt1 in a calcium dependent manner.
However ATP can only prevent the membrane binding of Syt1 if the PS concentration
is not higher than 15% and no PiP2 is present. Again, 15% PS is comparable to the
natural lipid composition in SVs and CGs. On the other hand, 1% PiP2 enhances the
calcium binding between Syt1 and the membrane by more than a factor of four, possibly due to the direct binding of the polybasic stretch of the C2B domain to PiP2 [89].
In conclusion, Syt1 binds the vesicular membrane in cis in the absence of ATP and
becomes inactive for cross-linking the plasma membrane, whereas ATP interrupts the
cis-binding of Syt1 and activates the C2 domains to bind the plasma membrane containing PiP2 [89]. This finding is important, because the shielding model of ATP may
balance the cis- and trans-membrane interaction of Syt1 and thus modulate the membrane fusion. Cross-linking could be thought of as an “intermediate” of both cis-and
trans-binding, so that the C2B domain interacts with PiP2 in the plasma membrane and
C2A backbinds to the vesicular membrane [141]. If this is the case, it is still not clear
whether the cross-linking proceeds in a single step or in a sequence of steps.
Moreover, the trans-binding of Syt1 already tethers the vesicle to the plasma membrane, since Syt1 is anchored in the vesicular membrane by the transmembrane domain, raising the question whether a calcium dependent cross-linking is still required
for its function in membrane fusion.
The advantage of a cross-linking model is to lower the distance between the membranes and promote the SNARE assembly [141]. Finally, it cannot be ruled out that calcium dependent cis-binding is sufficient for triggering the membrane fusion, in such
fashion as to induce change in the vesicular membrane curvature.
77
4.2
α-SNAP inhibits SNARE-mediated CG fusion, but
does not fully block CG docking
Neuronal exocytosis is mediated by SNAREs assembly, termed “zippering” of the
vesicular protein synaptobrevin and the Q-SNARE proteins syntaxin and SNAP-25 located in the target membrane [63]. The zippering proceeds from the N-terminus to the
C-terminus [91]. α-SNAP has been reported to inhibit vesicle fusion [4, 9, 103, 120, 127].
The preliminary fusion experiments between CGs and reconstituted liposome performed in the neurobiology department have shown that α-SNAP completely prevents
CG fusion and that the inhibition requires binding of the α1 region of the N-terminus
to the C-terminus of the Q-SNARE complex near the membrane and the membrane attachment of the hydrophobic loop of the N-terminus (unpublished data). As an extension of these results, docking between the purified CGs and the 100 nm liposome was
investigated in the presence of α-SNAP or its N-terminus mutants (α-SNAPdel and αSNAPF27,28S ). Liposomes contained stabilized Q-SNARE complex. Docking was monitored via diffusion time changing using FCS. Surprisingly, docking was still observable
under the same conditions, by which fusion is prevented. The negative control using
soluble Sb1−96 restored the docking, indicating that docking is SNARE dependent. In
conclusion, α-SNAP blocks the CG fusion at the C-terminal side of the SNARE complex
(unpublished data of the neurobiology department), but does not prevent the partial
SNARE assembly starting from the SNARE N-terminal end, and preserves the vesicle
in a docked intermediate.
This unexpected result has never been observed before and sheds light to the regulatory role of the α-SNAP in membrane fusion. However it is necessary to question
whether a partial assembly of SNARE proteins at the C-terminal side is conceivable.
The ternary SNARE complex structures has highly conserved layers, in which the side
chains of the amino acid interact in the “tube” of the four SNARE motives [36]. Furthermore, the X-ray study suggested that the SNARE helical bundle is extended from
the SNARE motives to the linker region by the further side chain interaction and the
78
linker region directly contributes to membrane fusion [112]. It has also been reported,
that the transition region between the SNARE motives and the linker region plays
an important role in membrane fusion [106, 109, 126]. Due to these, a single deletion
mutant of synaptobrevin was introduced at position 84 (Sb ∆84), which consists the
+8 layer of the SNARE helical bundle and is directly followed by the linker region [55].
Sb ∆84 bearing liposomes cannot fuse if the liposome contains stabilized Q-SNARE
complex (∆N complex) as long as at least one of the liposome populations is comprised of large liposomes (radius from 40 to 100 nm), since the +8 layer is disrupted by
deletion. However, the FCCS results combined with FRET lifetime analysis [23] have
shown that Sb ∆84 still docks the large liposome, indicating that partial SNARE assembly still proceeds under these conditions [55]. Although this result cannot be directly
compared with the α-SNAP mediated partial SNARE assembly, it suggests a scenario
in which SNARE assembly can be halted at the C-terminus end, if the layer structure
of the SNARE complex is disrupted. Thereby, it is conceivable that the interaction between the α1 region of α-SNAP and the SNARE’s C-terminal side near the membrane
may be strong enough to deform one of the layers and dock the liposomes by partial
SNARE assembly. Finally, it has to be pointed out, that the increased membrane curvature of the small liposome (diameter 45 nm) overcomes the inhibition of α-SNAP and
promotes the membrane fusion (unpublished data from Dr. Yongsoo Park).
79
4.3
Quantitative characterization of synaptic vesicles
Using a sensitive vesicle counting assay based on TP-FCS, freshly isolated homoge-
nous mouse synaptic vesicles with low density were characterized in terms of the particle number. The membrane labeling dye FM 1-43 was used to visualize the vesicles.
Cooperating with the neurobiology department, the total protein mass of the same
vesicle sample was determined using a modified Lowry-Peterson method. Combining
both results, an averaged protein mass per single synaptic vesicle was obtained.
The result is very well related to the previous report of rat synaptic vesicle [119],
indicating that the two different species have a similar property of the synaptic vesicles despite of their morphological differences in terms of body - and brain size. In
particular, mouse synaptic vesicles have a larger protein mass than those of obtained
from rats (∼ 20%). With the known vesicle number as a starting point, many biological
parameters can be identified at a single vesicle level, for instance protein composition
and lipid composition.
80
Chapter 5
Conclusions
Neurotransmission requires rapid fusion of synaptic vesicles with the presynaptic
membrane. This fusion is triggered by calcium influx and promoted by SNARE assembly. The vesicular protein synaptotagmin-1 acts as a calcium sensor and its transtethering to the plasma membrane is an essential step in the membrane fusion process.
In this study, several details in membrane fusion have been investigated and discussed.
First, the molecular requirements of the trans- and cis-membrane interaction of
synaptotagmin-1 have been analyzed. All of the three binding sites of synaptotagmin-1
(the C2A domain, the C2B domain and the polylysine patch) are involved in both of
these processes. Cis-binding to the host membrane containing 20% phosphatidylserine seems to be a stable state and can be overcome by neither trans-binding between
synaptotagmin-1 and the target membrane containing acidic phospholipids, nor by the
interaction between synaptotagmin-1 and SNAREs inserted into the acceptor membrane. Similarly, with 12% phosphatidylserine in the host membrane–a situation comparable to that in wild-type vesicles [119]–cis-binding is found nearly exclusively.
With 5% phosphatidylserine, however, the presence of either phosphatidylinositol-4,5bisphosphate in the target membrane or calcium in the solution can invoke conversion
to trans-binding with great efficiency.
Second, liposome cross-linking can be mediated by soluble C2AB fragments, when
81
CHAPTER 5. CONCLUSIONS
all the binding sites for C2AB are saturated on the liposome surface. To rationalize
this result, it is assumed that binding with only the C2A- or the C2B domain might
occur under these conditions. The other–unbound–domain of the same synaptotagmin
molecule will eventually lead to cross-linking, if it attaches to a different liposome. This
finding confirms the observation of cis-binding and explains some of the conflicts in
the existing literature. Apparently, a subtle balance between cis- and trans-binding of
synaptotagmin-1 may play an important role in the regulation of neuronal membrane
fusion.
Moreover, first data for the characterization of mouse synaptic vesicles, in terms of
average protein mass per vesicle, was measured in this study. The results obtained for
mouse synaptic vesicles are similar to those of rat synaptic vesicles and offer a single
vesicle basis for further investigating mouse synaptic vesicles as well as membrane
fusion in a quantitative fashion.
Finally, details of the inhibiting effect of α-SNAP on the fusion of chromaffin granules with larger liposomes were also investigated in the present study. The experimental findings suggest that α-SNAP inhibits fusion of chromaffin granules and stops
the SNARE assembly at the C-terminal site, so that partial SNARE zippering can still
mediate docking of chromaffin granules.
82
Bibliography
[1] A HMED , S. Isolation and Characterization of Synaptic Vesicles from Mouse Brain. PhD
thesis, Georg-August-University Göttingen, 2010.
[2] A NGLESON , J. K., C OCHILLA , A. J., K ILIC , G., N USSINOVITCH , I.,
AND
B ETZ ,
W. J. Regulation of dense core release from neuroendocrine cells revealed by
imaging single exocytic events. Nat Neurosci 2, 5 (May 1999), 440–446.
[3] A RAÇ , D., C HEN , X., K HANT, H. A., U BACH , J., L UDTKE , S. J., K IKKAWA ,
M., J OHNSON , A. E., C HIU , W., S ÜDHOF, T. C.,
AND
R IZO , J.
Close
membrane-membrane proximity induced by Ca2+ -dependent multivalent binding of synaptotagmin-1 to phospholipids. Nat Struct Mol Biol 13, 3 (Mar 2006),
209–217.
[4] B ABCOCK , M., M ACLEOD , G. T., L EITHER , J., AND PALLANCK , L. Genetic analysis of soluble N-ethylmaleimide-sensitive factor attachment protein function in
Drosophila reveals positive and negative secretory roles. J Neurosci 24, 16 (Apr
2004), 3964–3973.
[5] B ACIA , K.,
AND
S CHWILLE , P. Fluorescence correlation spectroscopy. Methods
Mol Biol 398 (2007), 73–84.
[6] B AI , J., T UCKER , W. C.,
AND
C HAPMAN , E. R. PIP2 increases the speed of
response of synaptotagmin and steers its membrane-penetration activity toward
the plasma membrane. Nat Struct Mol Biol 11, 1 (Jan 2004), 36–44.
83
BIBLIOGRAPHY
[7] B AI , J., WANG , C.-T., R ICHARDS , D. A., J ACKSON , M. B.,
AND
C HAPMAN ,
E. R. Fusion pore dynamics are regulated by synaptotagmin*t-SNARE interactions. Neuron 41, 6 (Mar 2004), 929–942.
[8] B ARNARD , R. J., M ORGAN , A.,
AND
B URGOYNE , R. D. Stimulation of NSF
ATPase activity by α-SNAP is required for SNARE complex disassembly and
exocytosis. J Cell Biol 139, 4 (Nov 1997), 875–883.
[9] B ARSZCZEWSKI , M., C HUA , J. J., S TEIN , A., W INTER , U., H EINTZMANN , R.,
Z ILLY, F. E., FASSHAUER , D., L ANG , T., AND J AHN , R. A novel site of action for
α-SNAP in the SNARE conformational cycle controlling membrane fusion. Mol
Biol Cell 19, 3 (Mar 2008), 776–784.
[10] B ERLAND , K. M., S O , P. T.,
AND
G RATTON , E. Two-photon fluorescence cor-
relation spectroscopy: method and application to the intracellular environment.
Biophys J 68, 2 (Feb 1995), 694–701.
[11] B ERNARD , V. Molecular Fluorescence: Principles and Applications. Wiley-VCH,
2001.
[12] B ETZ , W. J., M AO , F., AND B EWICK , G. S. Activity-dependent fluorescent staining and destaining of living vertebrate motor nerve terminals. J Neurosci 12, 2
(Feb 1992), 363–375.
[13] B HALLA , A., C HICKA , M. C., T UCKER , W. C.,
AND
C HAPMAN , E. R. Ca2+ -
synaptotagmin directly regulates t-SNARE function during reconstituted membrane fusion. Nat Struct Mol Biol 13, 4 (Apr 2006), 323–330.
[14] B LOM , H., C HMYROV, A., H ASSLER , K., D AVIS , L.,
AND
W IDENGREN , J.
Triplet-State Investigations of Fluorescent Dyes at Dielectric Interfaces Using Total Internal Reflection Fluorescence Correlation Spectroscopy. J Phys Chem A 113,
19 (2009), 5554–5566.
84
BIBLIOGRAPHY
[15] B RADFORD , M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal
Biochem 72 (May 1976), 248–254.
[16] B RUMBACK , A. C., L IEBER , J. L., A NGLESON , J. K.,
AND
B ETZ , W. J. Using
FM1-43 to study neuropeptide granule dynamics and exocytosis. Methods 33, 4
(Aug 2004), 287–294.
[17] B RUNGER , A. T. Structure of proteins involved in synaptic vesicle fusion in
neurons. Annu Rev Biophys Biomol Struct 30 (2001), 157–171.
[18] B URGALOSSI , A., J UNG , S., M EYER , G., J OCKUSCH , W. J., J AHN , O., TASCHEN BERGER ,
AND
H., O’C ONNOR , V. M., N ISHIKI , T.- I ., TAKAHASHI , M., B ROSE , N.,
R HEE , J.-S. SNARE protein recycling by αSNAP and βSNAP supports
synaptic vesicle priming. Neuron 68, 3 (Nov 2010), 473–487.
[19] C HAPMAN , E. R. How does synaptotagmin trigger neurotransmitter release?
Annu Rev Biochem 77 (2008), 615–641.
[20] C HEN , Y. A.,
AND
S CHELLER , R. H. SNARE-mediated membrane fusion. Nat
Rev Mol Cell Biol 2, 2 (Feb 2001), 98–106.
[21] C HOI , U. B., S TROP, P., V RLJIC , M., C HU , S., B RUNGER , A. T., AND W ENINGER ,
K. R. Single-molecule FRET-derived model of the synaptotagmin 1-SNARE fusion complex. Nat Struct Mol Biol 17, 3 (Mar 2010), 318–324.
[22] C LARY, D. O., G RIFF , I. C.,
AND
R OTHMAN , J. E. SNAPs, a family of NSF
attachment proteins involved in intracellular membrane fusion in animals and
yeast. Cell 61, 4 (May 1990), 709–721.
[23] C YPIONKA , A., S TEIN , A., H ERNANDEZ , J. M., H IPPCHEN , H., J AHN , R.,
AND
WALLA , P. J. Discrimination between docking and fusion of liposomes reconstituted with neuronal SNARE-proteins using FCS. Proc Natl Acad Sci U S A 106, 44
(Nov 2009), 18575–18580.
85
BIBLIOGRAPHY
[24] D AUTY, E.,
V ERKMAN , A. S. Actin cytoskeleton as the principal determi-
AND
nant of size-dependent DNA mobility in cytoplasm: a new barrier for non-viral
gene delivery. J Biol Chem 280, 9 (Mar 2005), 7823–7828.
[25] D AVIS , L. M.,
AND
S HEN , G. Accounting for triplet and saturation effects in
FCS measurements. Curr Pharm Biotechnol 7, 4 (Aug 2006), 287–301.
[26] D E C AMILLI , P.,
AND J AHN ,
R. Pathways to regulated exocytosis in neurons.
Annu Rev Physiol 52 (1990), 625–645.
[27] D ENK , W., S TRICKLER , J. H.,
AND
W EBB , W. W. Two-photon laser scanning
fluorescence microscopy. Science 248, 4951 (Apr 1990), 73–76.
[28] E DELHOCH , H. Spectroscopic determination of tryptophan and tyrosine in proteins. Biochemistry 6, 7 (Jul 1967), 1948–1954.
[29] E HRENBERG , M., AND R IGLER , R. Rotational brownian motion and fluorescence
intensify fluctuations. Chemical Physics 4, 3 (1974), 390–401.
[30] E INSTEIN , A.
Über die von der molekularkinetischen Theorie der Wärme
geforderte Bewegung von in ruhenden Flüssigkeiten suspendierten Teilchen.
Ann Phys 17 (1905), 549–560.
[31] E LSON , E.,
M AGDE , D. Fluorescence correlation spectroscopy I: Concep-
AND
tual basis and theory. Biopolymers 13 (1974), 1–27.
[32] E X PAS Y
–
B IOINFORMATICS
R ESOURCE
P ORTAL.
http://web.expasy.org/protparam/. Accessed on September 25, 2012.
[33] FASSHAUER , D., A NTONIN , W., M ARGITTAI , M., PABST, S.,
AND
J AHN , R.
Mixed and non-cognate SNARE complexes. Characterization of assembly and
biophysical properties. J Biol Chem 274, 22 (May 1999), 15440–15446.
[34] FASSHAUER , D.,
AND
M ARGITTAI , M. A transient N-terminal interaction of
SNAP-25 and syntaxin nucleates SNARE assembly. J Biol Chem 279, 9 (Feb 2004),
7613–7621.
86
BIBLIOGRAPHY
[35] FASSHAUER , D., O TTO , H., E LIASON , W. K., J AHN , R.,
AND
B R ÜNGER , A. T.
Structural changes are associated with soluble N-ethylmaleimide-sensitive fusion protein attachment protein receptor complex formation. J Biol Chem 272, 44
(Oct 1997), 28036–28041.
[36] FASSHAUER , D., S UTTON , R. B., B RUNGER , A. T.,
AND J AHN ,
R. Conserved
structural features of the synaptic fusion complex: SNARE proteins reclassified
as Q- and R-SNAREs. Proc Natl Acad Sci U S A 95, 26 (Dec 1998), 15781–15786.
[37] F ERNANDEZ , I., A RAÇ , D., U BACH , J., G ERBER , S. H., S HIN , O., G AO , Y., A N DERSON ,
R. G., S ÜDHOF, T. C.,
AND
R IZO , J. Three-dimensional structure of
the synaptotagmin 1 C2B-domain: synaptotagmin 1 as a phospholipid binding
machine. Neuron 32, 6 (Dec 2001), 1057–1069.
[38] F IEBIG , K. M., R ICE , L. M., P OLLOCK , E.,
AND
B RUNGER , A. T. Folding inter-
mediates of SNARE complex assembly. Nat Struct Biol 6, 2 (Feb 1999), 117–123.
[39] F LEMING , K. G., H OHL , T. M., Y U , R. C., M ÜLLER , S. A., W OLPENSINGER , B.,
E NGEL , A., E NGELHARDT, H., B R ÜNGER , A. T., S ÖLLNER , T. H.,
SON ,
AND
H AN -
P. I. A revised model for the oligomeric state of the N-ethylmaleimide-
sensitive fusion protein, NSF. J Biol Chem 273, 25 (Jun 1998), 15675–15681.
[40] F URST, J., S UTTON , R. B., C HEN , J., B RUNGER , A. T.,
AND
G RIGORIEFF , N.
Electron cryomicroscopy structure of N-ethyl maleimide sensitive factor at 11 Å
resolution. EMBO J 22, 17 (Sep 2003), 4365–4374.
[41] F ÖRSTER , T. Zwischenmolekulare Energiewanderung und Fluoreszenz. Ann
Phys 6 (1948), 55–75.
[42] G AFFANEY, J. D., D UNNING , F. M., WANG , Z., H UI , E.,
AND
C HAPMAN , E. R.
Synaptotagmin C2B domain regulates Ca2+ -triggered fusion in vitro: critical
residues revealed by scanning alanine mutagenesis. J Biol Chem 283, 46 (Nov
2008), 31763–31775.
87
BIBLIOGRAPHY
[43] G ILL , S. C.,
AND VON
H IPPEL , P. H. Calculation of protein extinction coeffi-
cients from amino acid sequence data. Anal Biochem 182, 2 (Nov 1989), 319–326.
[44] G OEPPERT-M AYER , M. Über Elementarakte mit zwei Quantensprüngen. Annalen der Physik 9 (1931), 273–294.
[45] G RANSETH , B., O DERMATT, B., R OYLE , S. J.,
AND
L AGNADO , L. Clathrin-
mediated endocytosis is the dominant mechanism of vesicle retrieval at hippocampal synapses. Neuron 51, 6 (Sep 2006), 773–786.
[46] G R ÜNWALD , D., C ARDOSO , M. C., L EONHARDT, H.,
AND
B USCHMANN , V.
Diffusion and binding properties investigated by Fluorescence Correlation Spectroscopy (FCS). Curr Pharm Biotechnol 6, 5 (Oct 2005), 381–386.
[47] G ÖSCH , M., AND R IGLER , R. Fluorescence correlation spectroscopy of molecular
motions and kinetics. Adv Drug Deliv Rev 57, 1 (Jan 2005), 169–190.
[48] H ANSON , P. I., H EUSER , J. E.,
AND J AHN ,
R. Neurotransmitter release - four
years of SNARE complexes. Curr Opin Neurobiol 7, 3 (Jun 1997), 310–315.
[49] H ANSON , P. I., R OTH , R., M ORISAKI , H., J AHN , R.,
AND
H EUSER , J. E. Struc-
ture and conformational changes in NSF and its membrane receptor complexes
visualized by quick-freeze/deep-etch electron microscopy. Cell 90, 3 (Aug 1997),
523–535.
[50] H AUSTEIN , E., AND S CHWILLE , P. Fluorescence correlation spectroscopy: novel
variations of an established technique. Annu Rev Biophys Biomol Struct 36 (2007),
151–169.
[51] H AYASHI , T., M C M AHON , H., YAMASAKI , S., B INZ , T., H ATA , Y., S ÜDHOF,
T. C.,
AND
N IEMANN , H. Synaptic vesicle membrane fusion complex: action of
clostridial neurotoxins on assembly. EMBO J 13, 21 (Nov 1994), 5051–5061.
88
BIBLIOGRAPHY
[52] H AYASHI , T., YAMASAKI , S., N AUENBURG , S., B INZ , T.,
AND
N IEMANN , H.
Disassembly of the reconstituted synaptic vesicle membrane fusion complex in
vitro. EMBO J 14, 10 (May 1995), 2317–2325.
[53] H ELL , J. W., M AYCOX , P. R., S TADLER , H.,
AND J AHN ,
R. Uptake of GABA by
rat brain synaptic vesicles isolated by a new procedure. EMBO J 7, 10 (Oct 1988),
3023–3029.
[54] H ENKEL , A. W., L ÜBKE , J.,
AND
B ETZ , W. J. FM1-43 dye ultrastructural local-
ization in and release from frog motor nerve terminals. Proc Natl Acad Sci U S A
93, 5 (Mar 1996), 1918–1923.
[55] H ERNANDEZ , J. M., S TEIN , A., B EHRMANN , E., R IEDEL , D., C YPIONKA , A.,
FARSI , Z., WALLA , P. J., R AUNSER , S.,
AND J AHN ,
R. Membrane fusion inter-
mediates via directional and full assembly of the SNARE complex. Science 336,
6088 (Jun 2012), 1581–1584.
[56] H ERRICK , D. Z., S TERBLING , S., R ASCH , K. A., H INDERLITER , A., AND C AFISO ,
D. S. Position of synaptotagmin I at the membrane interface: cooperative interactions of tandem C2 domains. Biochemistry 45, 32 (Aug 2006), 9668–9674.
[57] H OHL , T. M., PARLATI , F., W IMMER , C., R OTHMAN , J. E., S ÖLLNER , T. H.,
AND
E NGELHARDT, H. Arrangement of subunits in 20 S particles consisting of
NSF, SNAPs, and SNARE complexes. Mol Cell 2, 5 (Nov 1998), 539–548.
[58] H OLT, M., R IEDEL , D., S TEIN , A., S CHUETTE , C.,
AND
J AHN , R. Synaptic
vesicles are constitutively active fusion machines that function independently
of Ca2+ . Curr Biol 18, 10 (May 2008), 715–722.
[59] H UI , E., B AI , J., AND C HAPMAN , E. R. Ca2+ -triggered simultaneous membrane
penetration of the tandem C2-domains of synaptotagmin I. Biophys J 91, 5 (Sep
2006), 1767–1777.
89
BIBLIOGRAPHY
[60] H UI , E., G AFFANEY, J. D., WANG , Z., J OHNSON , C. P., E VANS , C. S.,
AND
C HAPMAN , E. R. Mechanism and function of synaptotagmin-mediated membrane apposition. Nat Struct Mol Biol 18, 7 (Jul 2011), 813–821.
[61] H UI , E., J OHNSON , C. P., YAO , J., D UNNING , F. M.,
AND
C HAPMAN , E. R.
Synaptotagmin-mediated bending of the target membrane is a critical step in
Ca2+ -regulated fusion. Cell 138, 4 (Aug 2009), 709–721.
[62] J AHN , R., L ANG , T.,
AND
S ÜDHOF, T. C. Membrane fusion. Cell 112, 4 (Feb
2003), 519–533.
[63] J AHN , R.,
AND
S CHELLER , R. H. SNAREs–engines for membrane fusion. Nat
Rev Mol Cell Biol 7, 9 (Sep 2006), 631–643.
[64] J AHN , R., S CHIEBLER , W.,
AND
G REENGARD , P.
A quantitative dot-
immunobinding assay for proteins using nitrocellulose membrane filters. Proc
Natl Acad Sci U S A 81, 6 (Mar 1984), 1684–1687.
[65] K IM , J.-Y., C HOI , B.-K., C HOI , M.-G., K IM , S.-A., L AI , Y., S HIN , Y.-K.,
AND
L EE , N. K. Solution single-vesicle assay reveals PIP2-mediated sequential actions of synaptotagmin-1 on SNAREs. EMBO J 31, 9 (May 2012), 2144–2155.
[66] K UNER , T., L I , Y., G EE , K. R., B ONEWALD , L. F., AND A UGUSTINE , G. J. Photolysis of a caged peptide reveals rapid action of N-ethylmaleimide sensitive factor
before neurotransmitter release. Proc Natl Acad Sci U S A 105, 1 (Jan 2008), 347–
352.
[67] K YOUNG , M., S RIVASTAVA , A., Z HANG , Y., D IAO , J., V RLJIC , M., G ROB , P.,
N OGALES , E., C HU , S.,
AND
B RUNGER , A. T. In vitro system capable of differ-
entiating fast Ca2+ -triggered content mixing from lipid exchange for mechanistic
studies of neurotransmitter release. Proc Natl Acad Sci U S A 108, 29 (Jul 2011),
E304–E313.
[68] L AEMMLI , U. K. Cleavage of structural proteins during the assembly of the head
of bacteriophage T4. Nature 227, 5259 (Aug 1970), 680–685.
90
BIBLIOGRAPHY
[69] L AI , A. L., H UANG , H., H ERRICK , D. Z., E PP, N., AND C AFISO , D. S. Synaptotagmin 1 and SNAREs form a complex that is structurally heterogeneous. J Mol
Biol 405, 3 (Jan 2011), 696–706.
[70] L AI , Y., AND S HIN , Y.-K. The importance of an asymmetric distribution of acidic
lipids for synaptotagmin 1 function as a Ca2+ sensor. Biochem J 443, 1 (Apr 2012),
223–229.
[71] L EE , H.-K., YANG , Y., S U , Z., H YEON , C., L EE , T.-S., L EE , H.-W., K WEON ,
D.-H., S HIN , Y.-K.,
AND
Y OON , T.-Y. Dynamic Ca2+ -dependent stimulation of
vesicle fusion by membrane-anchored synaptotagmin 1. Science 328, 5979 (May
2010), 760–763.
[72] L EUNG ,
Molecular
K.
Galactosyl
Imaging
and
serum
Contrast
albumin-rhodamine
Agent
Database
http://www.ncbi.nlm.nih.gov/books/NBK23508/, entry created:
green.
(MICAD):
Novem-
ber 26, 2007 (Updated January 15, 2008).
[73] L I , L., S HIN , O.-H., R HEE , J.-S., A RAÇ , D., R AH , J.-C., R IZO , J., S ÜDHOF, T.,
AND
R OSENMUND , C. Phosphatidylinositol phosphates as co-activators of Ca2+
binding to C2 domains of synaptotagmin 1. J Biol Chem 281, 23 (Jun 2006), 15845–
15852.
[74] M ADGE , D., E LSON , E.,
AND
W EBB , W. Thermodynamic Fluctuations in a Re-
acting System - Measurement by Fluorescence Correlation Spectroscopy. Phys.
Rev. Lett. 29 (1972), 705–708.
[75] M AGDE , D.,
AND
E LSON , E. L. Fluorescence Correlation Spectroscopy. III. Uni-
form Translation and Laminar Flow. Biopolymers 17 (1978), 361–376.
[76] M AGDE , D., E LSON , E. L.,
AND
W EBB , W. W. Fluorescence correlation spec-
troscopy. II. An experimental realization. Biopolymers 13, 1 (Jan 1974), 29–61.
[77] M ALHOTRA , V., O RCI , L., G LICK , B. S., B LOCK , M. R.,
AND
R OTHMAN , J. E.
Role of an N-ethylmaleimide-sensitive transport component in promoting fusion
91
BIBLIOGRAPHY
of transport vesicles with cisternae of the Golgi stack. Cell 54, 2 (Jul 1988), 221–
227.
[78] M ARTENS , S., K OZLOV, M. M.,
AND
M C M AHON , H. T. How synaptotagmin
promotes membrane fusion. Science 316, 5828 (May 2007), 1205–1208.
[79] M ARTENS , S.,
AND
M C M AHON , H. T. Mechanisms of membrane fusion: dis-
parate players and common principles. Nat Rev Mol Cell Biol 9, 7 (Jul 2008),
543–556.
[80] M ARZ , K. E., L AUER , J. M.,
AND
H ANSON , P. I. Defining the SNARE complex
binding surface of α-SNAP: implications for SNARE complex disassembly. J Biol
Chem 278, 29 (Jul 2003), 27000–27008.
[81] M AYER , A., W ICKNER , W.,
AND
H AAS , A. Sec18p (NSF)-driven release of
Sec17p (α-SNAP) can precede docking and fusion of yeast vacuoles. Cell 85, 1
(Apr 1996), 83–94.
[82] M C M AHON , H. T., K OZLOV, M. M.,
AND
M ARTENS , S. Membrane curvature
in synaptic vesicle fusion and beyond. Cell 140, 5 (Mar 2010), 601–605.
[83] M ÜLLER , M.,
AND
D AVIS , G. W. Vesicle priming in a SNAP. Neuron 68, 3 (Nov
2010), 324–326.
[84] N AGIEC , E. E., B ERNSTEIN , A.,
AND
W HITEHEART, S. W. Each domain of the
N-ethylmaleimide-sensitive fusion protein contributes to its transport activity. J
Biol Chem 270, 49 (Dec 1995), 29182–29188.
[85] N AGY, A., B AKER , R. R., M ORRIS , S. J., AND W HITTAKER , V. P. The preparation
and characterization of synaptic vesicles of high purity. Brain Res 109, 2 (Jun
1976), 285–309.
[86] N EHER , E. A comparison between exocytic control mechanisms in adrenal chromaffin cells and a glutamatergic synapse. Pflugers Arch 453, 3 (Dec 2006), 261–
268.
92
BIBLIOGRAPHY
[87] O TTO , H., H ANSON , P. I., AND J AHN , R. Assembly and disassembly of a ternary
complex of synaptobrevin, syntaxin, and SNAP-25 in the membrane of synaptic
vesicles. Proc Natl Acad Sci U S A 94, 12 (Jun 1997), 6197–6201.
[88] PACE , C. N., VAJDOS , F., F EE , L., G RIMSLEY, G., AND G RAY, T. How to measure
and predict the molar absorption coefficient of a protein. Protein Sci 4, 11 (Nov
1995), 2411–2423.
[89] PARK , Y., H ERNANDEZ , J. M.,
R IEDEL , D.,
AND J AHN ,
VAN DEN
B OGAART , G., A HMED , S., H OLT, M.,
R. Controlling synaptotagmin activity by electrostatic
screening. Nat Struct Mol Biol (Sep 2012), http://dx.doi.org/10.1038/nsmb.2375.
[90] P ETERSON , G. L. A simplification of the protein assay method of Lowry et al.
which is more generally applicable. Anal Biochem 83, 2 (Dec 1977), 346–356.
[91] P OBBATI , A. V., S TEIN , A., AND FASSHAUER , D. N- to C-terminal SNARE complex assembly promotes rapid membrane fusion. Science 313, 5787 (Aug 2006),
673–676.
[92] P OHL , W. H.
Entwicklung und Anwendung spektroskopischer Korrelationsver-
fahren zur Beantwortung biomolekularer Fragestellungen. PhD thesis, Georg-AugustUniversity Göttingen, 2009.
[93] P OHL , W. H., H ELLMUTH , H., H ILBERT, M., S EIBEL , J.,
AND
WALLA , P. J. A
two-photon fluorescence-correlation study of lectins interacting with carbohydrated 20 nm beads. Chembiochem 7, 2 (Feb 2006), 268–274.
[94] P ÜSCHEL , A. W., O’C ONNOR , V.,
AND
B ETZ , H.
The N-ethylmaleimide-
sensitive fusion protein (NSF) is preferentially expressed in the nervous system.
FEBS Lett 347, 1 (Jun 1994), 55–58.
[95] R ADHAKRISHNAN , A., S TEIN , A., J AHN , R.,
AND
FASSHAUER , D. The Ca2+
affinity of synaptotagmin 1 is markedly increased by a specific interaction of its
C2B domain with phosphatidylinositol 4,5-bisphosphate. J Biol Chem 284, 38 (Sep
2009), 25749–25760.
93
BIBLIOGRAPHY
[96] R HEE , J.-S., L I , L. Y., S HIN , O.-H., R AH , J.-C., R IZO , J., S ÜDHOF, T. C.,
AND
R OSENMUND , C. Augmenting neurotransmitter release by enhancing the apparent Ca2+ affinity of synaptotagmin 1. Proc Natl Acad Sci U S A 102, 51 (Dec 2005),
18664–18669.
[97] R ICHARDS , D. A., B AI , J.,
AND
C HAPMAN , E. R. Two modes of exocytosis at
hippocampal synapses revealed by rate of FM1-43 efflux from individual vesicles. J Cell Biol 168, 6 (Mar 2005), 929–939.
[98] R IGLER , R.,
AND
E LSON , E. Fluorescence Correlation Spectroscopy: Theory and Ap-
plications. Springer, 2001.
[99] R IGLER , R., F ÖLDES -PAPP, Z., M EYER -A LMES , F. J., S AMMET, C., V ÖLCKER ,
M.,
AND
S CHNETZ , A. Fluorescence cross-correlation: a new concept for poly-
merase chain reaction. J Biotechnol 63, 2 (Aug 1998), 97–109.
[100] R IGLER , R., M ETS , U., W IDENGREN , J.,
AND
K ASK , P. Fluorescence correlation
spectroscopy with high count rate and low background: analysis of translational
diffusion. Eur Biophys J 22 (1993), 169–175.
[101] R IZO , J., C HEN , X.,
AND
A RAÇ , D. Unraveling the mechanisms of synaptotag-
min and SNARE function in neurotransmitter release. Trends Cell Biol 16, 7 (Jul
2006), 339–350.
[102] R IZO , J., AND S ÜDHOF, T. C. Snares and Munc18 in synaptic vesicle fusion. Nat
Rev Neurosci 3, 8 (Aug 2002), 641–653.
[103] R ODR ÍGUEZ , F., B USTOS , M. A., Z ANETTI , M. N., R UETE , M. C., M AYORGA ,
L. S.,
AND
T OMES , C. N.
α-SNAP prevents docking of the acrosome dur-
ing sperm exocytosis because it sequesters monomeric syntaxin. PLoS One 6,
7 (2011), e21925.
[104] S CHIAVO , G., G U , Q. M., P RESTWICH , G. D., S ÖLLNER , T. H., AND R OTHMAN ,
J. E. Calcium-dependent switching of the specificity of phosphoinositide binding to synaptotagmin. Proc Natl Acad Sci U S A 93, 23 (Nov 1996), 13327–13332.
94
BIBLIOGRAPHY
[105] S CHUETTE , C. G., H ATSUZAWA , K., M ARGITTAI , M., S TEIN , A., R IEDEL , D.,
K ÜSTER , P., K ÖNIG , M., S EIDEL , C.,
AND J AHN ,
R. Determinants of liposome
fusion mediated by synaptic SNARE proteins. Proc Natl Acad Sci U S A 101, 9
(Mar 2004), 2858–2863.
[106] S CHWARTZ , M. L.,
AND
M ERZ , A. J. Capture and release of partially zipped
trans-SNARE complexes on intact organelles. J Cell Biol 185, 3 (May 2009), 535–
549.
[107] S CHWILLE , P., M EYER -A LMES , F. J.,
AND
R IGLER , R. Dual-color fluorescence
cross-correlation spectroscopy for multicomponent diffusional analysis in solution. Biophys J 72, 4 (Apr 1997), 1878–1886.
[108] S CH ÄGGER , H.,
AND VON
J AGOW , G.
Tricine-sodium dodecyl sulfate-
polyacrylamide gel electrophoresis for the separation of proteins in the range
from 1 to 100 kDa. Anal Biochem 166, 2 (Nov 1987), 368–379.
[109] S IDDIQUI , T. J., V ITES , O., S TEIN , A., H EINTZMANN , R., J AHN , R.,
AND
FASSHAUER , D. Determinants of synaptobrevin regulation in membranes. Mol
Biol Cell 18, 6 (Jun 2007), 2037–2046.
[110] S TEEL , G. J., B UCHHEIM , G., E DWARDSON , J. M.,
AND
W OODMAN , P. G. Evi-
dence for interaction of the fusion protein α-SNAP with membrane lipid. Biochem
J 325 ( Pt 2) (Jul 1997), 511–518.
[111] S TEIN , A., R ADHAKRISHNAN , A., R IEDEL , D., FASSHAUER , D.,
AND J AHN ,
R.
Synaptotagmin activates membrane fusion through a Ca2+ -dependent trans interaction with phospholipids. Nat Struct Mol Biol 14, 10 (Oct 2007), 904–911.
[112] S TEIN , A., W EBER , G., WAHL , M. C.,
AND J AHN ,
R. Helical extension of the
neuronal SNARE complex into the membrane. Nature 460, 7254 (Jul 2009), 525–
528.
[113] S UDHOF, T. C. The synaptic vesicle cycle. Annu Rev Neurosci 27 (2004), 509–547.
95
BIBLIOGRAPHY
[114] S UTTON , R. B., FASSHAUER , D., J AHN , R.,
AND
B RUNGER , A. T. Crystal struc-
ture of a SNARE complex involved in synaptic exocytosis at 2.4 A resolution.
Nature 395, 6700 (Sep 1998), 347–353.
[115] S ÖLLNER , T., B ENNETT, M. K., W HITEHEART, S. W., S CHELLER , R. H.,
AND
R OTHMAN , J. E. A protein assembly-disassembly pathway in vitro that may
correspond to sequential steps of synaptic vesicle docking, activation, and fusion. Cell 75, 3 (Nov 1993), 409–418.
[116] S ØRENSEN , J. B., W IEDERHOLD , K., M ÜLLER , E. M., M ILOSEVIC , I., N AGY,
G.,
DE
G ROOT , B. L., G RUBM ÜLLER , H.,
AND
FASSHAUER , D. Sequential N-
to C-terminal SNARE complex assembly drives priming and fusion of secretory
vesicles. EMBO J 25, 5 (Mar 2006), 955–966.
[117] S ÜDHOF, T. C. The synaptic vesicle cycle: a cascade of protein-protein interactions. Nature 375, 6533 (Jun 1995), 645–653.
[118] TAGAYA , M., W ILSON , D. W., B RUNNER , M., A RANGO , N.,
AND
R OTHMAN ,
J. E. Domain structure of an N-ethylmaleimide-sensitive fusion protein involved
in vesicular transport. J Biol Chem 268, 4 (Feb 1993), 2662–2666.
[119] TAKAMORI , S., H OLT, M., S TENIUS , K., L EMKE , E. A., G RØNBORG , M., R IEDEL ,
D., U RLAUB , H., S CHENCK , S., B R ÜGGER , B., R INGLER , P., M ÜLLER , S. A.,
R AMMNER , B., G R ÄTER , F., H UB , J. S., D E G ROOT , B. L., M IESKES , G.,
M ORIYAMA , Y., K LINGAUF, J., G RUBM ÜLLER , H., H EUSER , J., W IELAND , F.,
AND J AHN ,
R. Molecular anatomy of a trafficking organelle. Cell 127, 4 (Nov
2006), 831–846.
[120] T OMES , C. N., D E B LAS , G. A., M ICHAUT, M. A., FARR É , E. V., C HERHITIN ,
O., V ISCONTI , P. E.,
AND
M AYORGA , L. S. α-SNAP and NSF are required in a
priming step during the human sperm acrosome reaction. Mol Hum Reprod 11, 1
(Jan 2005), 43–51.
96
BIBLIOGRAPHY
[121] T ORRES , T., AND L EVITUS , M. Measuring conformational dynamics: a new FCSFRET approach. J Phys Chem B 111, 25 (Jun 2007), 7392–7400.
[122] VAN
DER
M EER , W., C OKER , G.,
AND
C HEN , S.-Y. Resonance Energy Transfer:
Theory and Data. Wiley VCH, 1994.
[123] V ENNEKATE , W., S CHR ÖDER , S., L IN , C.-C.,
WALD ,
M., J AHN , R.,
AND
VAN DEN
B OGAART, G., G RUN -
WALLA , P. J. Cis- and trans-membrane interactions
of synaptotagmin-1. Proc Natl Acad Sci U S A 109, 27 (Jul 2012), 11037–11042.
[124] V ERSTREKEN , P., O HYAMA , T., AND B ELLEN , H. J. FM 1-43 labeling of synaptic
vesicle pools at the Drosophila neuromuscular junction. Methods Mol Biol 440
(2008), 349–369.
[125] V RLJIC , M., S TROP, P., E RNST, J. A., S UTTON , R. B., C HU , S.,
AND
B RUNGER ,
A. T. Molecular mechanism of the synaptotagmin-SNARE interaction in Ca2+ triggered vesicle fusion. Nat Struct Mol Biol 17, 3 (Mar 2010), 325–331.
[126] WALTER , A. M., W IEDERHOLD , K., B RUNS , D., FASSHAUER , D.,
AND
S ØRENSEN , J. B. Synaptobrevin N-terminally bound to syntaxin-SNAP-25 defines the primed vesicle state in regulated exocytosis. J Cell Biol 188, 3 (Feb 2010),
401–413.
[127] WANG , L., U NGERMANN , C.,
AND
W ICKNER , W. The docking of primed vac-
uoles can be reversibly arrested by excess Sec17p (α-SNAP). J Biol Chem 275, 30
(Jul 2000), 22862–22867.
[128] W EBER , T., Z EMELMAN , B. V., M C N EW, J. A., W ESTERMANN , B., G MACHL ,
M., PARLATI , F., S ÖLLNER , T. H.,
AND
R OTHMAN , J. E. SNAREpins: minimal
machinery for membrane fusion. Cell 92, 6 (Mar 1998), 759–772.
[129] W IDENGREN , J., S CHWEINBERGER , E., B ERGER , S.,
AND
S EIDEL , C. Two New
Concepts to Measure Fluorescence Resonance Energy Transfer via Fluorescence
Correlation Spectroscopy: Theory and Experimental Realizations. The Journal of
Physical Chemistry A 105, 28 (2001), 6851–6866.
97
BIBLIOGRAPHY
[130] W IDENGREN , J.,
AND
S CHWILLE , P. Characterization of Photoinduced Isomer-
ization and Back-Isomerization of the Cyanine Dye Cy5 by Fluorescence Correlation Spectroscopy. The Journal of Physical Chemistry A 104, 27 (2000), 6416–6428.
[131] W IMMER , C., H OHL , T. M., H UGHES , C. A., M ÜLLER , S. A., S ÖLLNER , T. H.,
E NGEL , A., AND R OTHMAN , J. E. Molecular mass, stoichiometry, and assembly
of 20 S particles. J Biol Chem 276, 31 (Aug 2001), 29091–29097.
[132] W INTER , U., C HEN , X.,
AND
FASSHAUER , D. A conserved membrane attach-
ment site in α-SNAP facilitates N-ethylmaleimide-sensitive factor (NSF)-driven
SNARE complex disassembly. J Biol Chem 284, 46 (Nov 2009), 31817–31826.
[133] X U , T., A SHERY, U., B URGOYNE , R. D.,
AND
N EHER , E. Early requirement for
α-SNAP and NSF in the secretory cascade in chromaffin cells. EMBO J 18, 12 (Jun
1999), 3293–3304.
[134] X UE , M., M A , C., C RAIG , T. K., R OSENMUND , C.,
AND
R IZO , J. The Janus-
faced nature of the C(2)B domain is fundamental for synaptotagmin-1 function.
Nat Struct Mol Biol 15, 11 (Nov 2008), 1160–1168.
[135] YANG , X., K AESER -W OO , Y. J., PANG , Z. P., X U , W.,
AND
S ÜDHOF, T. C. Com-
plexin clamps asynchronous release by blocking a secondary Ca2+ sensor via its
accessory α helix. Neuron 68, 5 (Dec 2010), 907–920.
[136] Y U , R. C., J AHN , R.,
AND
B RUNGER , A. T. NSF N-terminal domain crystal
structure: models of NSF function. Mol Cell 4, 1 (Jul 1999), 97–107.
[137] Z HAO , C., S MITH , E. C.,
AND
W HITEHEART, S. W. Requirements for the cat-
alytic cycle of the N-ethylmaleimide-Sensitive Factor (NSF). Biochim Biophys Acta
1823, 1 (Jan 2012), 159–171.
[138]
VAN DEN
AND
B OGAART , G., H OLT, M. G., B UNT, G., R IEDEL , D., W OUTERS , F. S.,
J AHN , R. One SNARE complex is sufficient for membrane fusion. Nat
Struct Mol Biol 17, 3 (Mar 2010), 358–364.
98
BIBLIOGRAPHY
[139]
VAN DEN
B OGAART , G., AND J AHN , R. Inside insight to membrane fusion. Proc
Natl Acad Sci U S A 108, 29 (Jul 2011), 11729–11730.
[140]
VAN DEN
B OGAART , G., M EYENBERG , K., D IEDERICHSEN , U.,
AND J AHN ,
R.
Phosphatidylinositol 4,5-bisphosphate increases Ca2+ affinity of synaptotagmin1 by 40-fold. J Biol Chem 287, 20 (May 2012), 16447–16453.
[141]
VAN DEN
B OGAART , G., T HUTUPALLI , S., R ISSELADA , J. H., M EYENBERG , K.,
H OLT, M., R IEDEL , D., D IEDERICHSEN , U., H ERMINGHAUS , S., G RUBM ÜLLER ,
H., AND J AHN , R. Synaptotagmin-1 may be a distance regulator acting upstream
of SNARE nucleation. Nat Struct Mol Biol 18, 7 (Jul 2011), 805–812.
99
BIBLIOGRAPHY
100
List of Figures
1.1
Synaptic vesicle cycle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2
1.2
Four-helix-bundle of the core SNARE complex. . . . . . . . . . . . . . . .
3
1.3
Layer structure of the ternary SNARE complex. . . . . . . . . . . . . . . .
4
1.4
The SNARE conformational cycle during vesicle docking and fusion. . .
5
1.5
3D structure of synaptotagmin-1. . . . . . . . . . . . . . . . . . . . . . . .
8
1.6
Membrane attachment site of α-SNAP in the presence of the SNARE
complex. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
1.7
Fluorescence autocorrelation spectroscopy. . . . . . . . . . . . . . . . . .
12
1.8
Fluorescence cross-correlation-spectroscopy. . . . . . . . . . . . . . . . .
15
1.9
OPE compared with TPE in terms of the excitation module and the effective volume. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
17
1.10 Jablonski-diagram of FRET donor and acceptor. . . . . . . . . . . . . . .
18
1.11 Excitation and emission spectra of TR-DHPE and OG-DHPE. . . . . . . .
19
2.1
Expression test of Syt1 under different conditions. . . . . . . . . . . . . .
31
2.2
Purification of Syt1 with immobilized metal ion affinity chromatography and ion exchange chromatography. . . . . . . . . . . . . . . . . . . .
2.3
33
Size exclusion chromatography using sephadex G50 econo column to
reconstitute small liposomes. . . . . . . . . . . . . . . . . . . . . . . . . .
101
37
LIST OF FIGURES
2.4
Confocal fluorescence microscopy setup for two-photon excitation. . . .
41
3.1
Schematic of typical FCCS curves in the tethering assay. . . . . . . . . . .
44
3.2
Example FCCS data from the tethering experiments. . . . . . . . . . . . .
45
3.3
Domain structures of Syt1 and its mutants used in this experiment. . . .
46
3.4
Liposome tethering mediated by membrane-anchored Syt1. . . . . . . .
48
3.5
Syt1 mediated membrane tethering in the presence of 12% and of 5% PS
in the cis-membrane. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6
50
Domain structures of the soluble C2AB domain and its Alexa Fluor 488
labeled mutant. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
50
3.7
Clustering of liposomes mediated by soluble C2AB domain. . . . . . . .
51
3.8
Binding assay of soluble C2AB labeled with alexa fluor 488 to TRPS li-
3.9
posome containing 20% PS. . . . . . . . . . . . . . . . . . . . . . . . . . .
52
Binding kinetics between liposomes (A) and AF-C2AB (B). . . . . . . . .
54
3.10 Tethering of liposomes mediated by membrane-anchored Syt1-SNAREs
interactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
57
3.11 Molecular model of an average SV. . . . . . . . . . . . . . . . . . . . . . .
58
3.12 SV labeling with FM 1-43 dye. . . . . . . . . . . . . . . . . . . . . . . . . .
60
3.13 FCS measurement of rat SVs in three dilutions compared to FM 1-43
background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
61
3.14 SV concentration plotted versus the protein concentration of the same
sample. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
63
3.15 Calculated protein mass per SV for mouse compared to rat. . . . . . . . .
64
3.16 Schmatic of CG docking with 100 nm liposomes in the presence of α-SNAP. 66
3.17 Diffusion time determination of the liposome docking experiments with
CGs using FCS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
102
68
LIST OF FIGURES
3.18 Histogram of the docking experiment using liposome and CG. . . . . . .
69
3.19 α-SNAP does not block the SNARE-mediated CG docking. . . . . . . . .
71
4.1
Model of the trans-tethering of the membrane anchored Syt1 to the target
membrane containing either acidic phospholipids or SNAREs. . . . . . .
74
4.2
All trans-tethering is inhibited if the host liposome of Syt1 contains 20% PS. 75
4.3
C2AB fragment is able to cluster the liposome under a saturating concentration on the liposome surface. . . . . . . . . . . . . . . . . . . . . . .
103
75
LIST OF FIGURES
104
List of Tables
2.1
Phospholipids used for reconstitution of liposomes. . . . . . . . . . . . .
22
2.2
Proteins used in this study. . . . . . . . . . . . . . . . . . . . . . . . . . . .
24
2.3
Buffers used in this study . . . . . . . . . . . . . . . . . . . . . . . . . . .
25
2.4
Instruments, filters, columns and miscellanea. . . . . . . . . . . . . . . .
29
2.5
Tricine (Schägger)-gels. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
36
2.6
Lipid composition of the liposomes in the Syt1 tethering experiments. .
38
3.1
Binding constant between liposomes and AF-C2AB fragments. . . . . . .
53
105
LIST OF TABLES
106
Abbreviations and Symbols
α-SNAP
N-Ethylmaleimide-sensitive factor attachment protein, α
Abs
Absorption
aa
Amino acids
AAA
ATPases associated with a variety of cellular activities
APS
Ammonium persulfate
BSA
Bovine serum albumin
CG
Chromaffin granules
CHAPS
3-[(3-Cholamidopropyl)dimethylammonio]-1-propanesulfonate
Cys
Cysteine
ddH2 O
Double distilled water
DNAse I
Deoxyribonuclease I
DOC
Deoxycholic acid
DTT
Dithiothreitol
EDTA
Ethylenediaminetetraacetic acid
f.c.
Final concentration
fl
Femtoliter
107
ABBREVIATIONS AND SYMBOLS
FRET
Förster Resonance Energy Transfer
h
Hours
HEPES
4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid
IEXC
Ion Exchange Chromatography
IMAC
Immobilized Metal ion Affinity Chromatography
IPTG
Isopropyl-β-D-thiogalactopyranosid
KCl
Potassium chloride
LUVs
Large unilamellar vesicles
Milli-Q
Ultrapure water
min
Minutes
NA
Numerical aperture
NBD-DOPE
1,2-Dioleoyl-sn-glycero-3-phosphoethanolamine-N-(7-nitrobenz2-oxa-1,3-diazol-4-yl)
Ni-NTA
Ni2+ -nitrilotriacetic acid beads
NSF
N-Ethylmaleimide-sensitive factor
OD
Optical density
OGPE
Oregon Green-PE
OPE
One-photon excitation
P:L
Molar protein to lipid ratio
PC
L-α-Phosphatidylcholine
PE
L-α-Phosphatidylethanolamine
108
PI
L-α-Phosphatidylinositol
PiP2
Phosphatidylinositol-4,5-bisphosphate
PMSF
Phenylmethanesulfonylfluoride
PS
L-α-Phosphatidylserine
RG
Rhodamine Green
Rhodamine-DOPE
1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-lissamine rhodamine B sulfonyl ammonium salt
RPM
Revolutions per minute
RT
Room temperature
SDS
Sodium dodecyl sulfate
SDS-PAGE
Sodium dodecyl sulfate polyacrylamide gel electrophoresis
SEM
Standard error of the mean
SNAP25A
Synaptosomal-associated protein 25A
SNAPs
Soluble NSF attachment proteins
SNARE
Soluble N-ethylmaleimide-sensitive-factor attachment receptor
Sx1A
Syntaxin 1A
Syt1
Synaptotagmin-1
TCA
Trichloroacetic acid
TEMED
N,N,N’,N’-Tetramethylethylendiamine
TPE
Two-photon excitation
Tricine
N-(Tri(hydroxymethyl)methyl)glycine
109
Trp
Tryptophan
TRPE
Texas Red-PE
Tyr
Tyrosine
v/v
volume/volume
w/v
weight/volume
110
Publication
Vennekate W, Schröder S, Lin CC, van den Bogaart G, Grunwald M, Jahn R,
Walla PJ. “Cis- and trans-membrane interactions of synaptotagmin-1.” Proc. Natl.
Acad. Sci. USA 109, 11037–11042 (2012).
Another manuscript is in preparation.
111
Acknowledgement
I would like to thank both of my supervisors Professor Peter Jomo Walla and Professor Reinhard Jahn, for supporting me to complete my PhD thesis with many pleasant discussions and many important suggestions. Also I would like to thank Professor
Claudia Steinem for joining my thesis committee and giving me lots of useful suggestions.
My collaboration partners, Dr. Geert van den Bogaart, Dr. Yongsoo Park, Dr. Saheeb Ahmed and Dr. Matthew Holt, have to be named not only for offering an active
working atmosphere, but also for the growing friendships with them during the past
four years.
I would like to give special thanks to Dr. Anna Cypionka and Dr. Wiebke Hanna
Pohl for getting me start in the work group and sharing a great time with me.
I would like to thank all my friends of the Walla group and the Jahn department for
having an enjoyable time with them. I thank for the great help from Dr. Javier Matias
Hernandez, Jianhua Chen, Matthias Grunwald, Sabrina Schröder, Chao-Chen Lin and
Mrs. Ursel Ries during my pregnancy leave, which enabled me to continue my project
during an important period.
Finally, I would like to thank Inge Dreger and Hendrik Vennekate for revising the
manuscript.
Curriculum Vitae
Personal information
Name: Wensi Vennekate, née Gao
Date of birth: December 23rd, 1982
City of birth: Tongliaoshi, China
Nationality: German
Education
2009/01-present
Ph.D. student in Chemistry in the workgroup Biomolecular Spectroscopy and Single-Molecule Detection, Max-Planck-Institute for
Biophysical Chemistry, Göttingen, Germany.
2008/11-2008/12
Practical training in the workgroup Biomolecular Spectroscopy
and Single-Molecule Detection, Max-Planck-Institute for Biophysical Chemistry, Göttingen, Germany.
2003/10-2008/10
Diploma in Chemistry. Georg-August-University, Göttingen, Germany.
Thesis Title: Wechselwirkung neuer Cholinkinase-Inhibiroren mit den
Enzymen von Malaria-Plasmodium (Interaction of Novel Cholinkinase Inhibitors with Malaria Plasmodium).
2002/12-2003/09
German language course (DSH), Goethe-Institute, Göttingen, Germany.
2001/09-2002/08
Student of industrial design, Beijing University of Technology,
China.
1998/09-2001/08
Beijing No. 4 High School, China.

Investigation of Neuronal Membrane Fusion Using

Transcrição

Documentos relacionados

CONGRESS INFORMATION Committees

aparatus golgi

項目詳細信息 Detalhes de Projeto

Apresentação do PowerPoint

The role of polysaccharides on the grape must ultrafiltration

Pokemon fusion name generator

A. HALBREICH, J. ROGER, J.N. PONS, D. GELDWERTH, M.F. DA

S - The Fluorescence Foundation

Permanent hair removal of white, grey and light

Titanium Membranes in Prevention of Alveolar Collapse - ID-SC